-------
METSCALARS
determination of stability, however, since EXPCLASS and the wind speed are also
used.) In addition, the PTSRCE preprocessor, used to calculate point source emis-
sions, may use the temperature gradients in determining the effective stack height
of point source emissions.
Representation of the typically complex vertical structure of the atmosphere with
just two temperature gradients demands some compromise in determining their
values. Figures 6-11 shows idealized vertical temperature profiles for (1) a typical
early morning profile with two.distinct layers: a surface-based inversion and a near-
neutral layer above (Figure 6-1 la); (2) a mid-morning profile with a mixed layer near
the surface, a stable inversion layer, and a near neutral layer above (Figure 6-1 Ib);
and (3) an afternoon profile with an unstable layer in the lower part of the profile
with a neutral layer above (Figure 6-ilc). In each case a siigntly different method is
employed to determine the temperature gradients.
For the early morning profile (Figure 6-1 la) DIFFBREAK nas been set at the height
of the top of the temperature inversion (300 meters in the example); therefore
TGRADBELOW :s just the temperature gradient from the ground to :he diffusion
break, yielding a stable value of 0.017 K/m. Similarly, TGRADABOVE is tne
gradient from the diffusion break to the -egion top. which In this case is almost neu-
tral.
For the midmorning profile (Figure 6-1 Ib) TGRADBELOW is also the gradient from
the ground to the diffusion break, although in this case it is slightly unstable, -.014
K/m. For TGRADABOVE, the recommended value is the average gradient from the
diffusion break to the top of the region. Although this does not precisely represent
the gradients within or above the inversion, it is the best compromise. Some calcula-
ted plume rise may be too low because of this choice, but it is necessary to avoid
unrealistically high temperatures, which could be calculated if the positive gradient
were extended to several hundred meters above the top of the inversion. In this
example, TGRADABOVE would be -0.005 K/m.
For the afternoon profile (Figure 6-lie) the gradients are again relatively simple to
determine. TGRADBELOW is the somewhat unstable gradient of -0.0103 K/m from
the ground to tne diffusion ^reak. -vhile ""CRAD ABOVE 's again -ear!- neutral.
90008 19
229
-------
1000
800-
h
i
^ soof-
0)
.2?
5;
- 400
i
j-
i
200h
j I
Region Top
Diffusion
break
i i
290 295
Temperature (degrees K)
300
FIGURE 6-lla. Idealized early morning terrperature profile showing
a surface based inversion.
230
-------
METSCALARS
1000
800-
73
*-*
0)
600-
.2?
= 400
200H
§85
Diffusion
break
290 295
Temperature (degrees K)
300
FIGJTJE 6-l]b. Idealized nid morning tenperature profile showing
an elevated inversion.
231
-------
1000
I
800-
71
«
0>
500
.2?
= 400h
200
J85
Region .Top
Diffusion
break
290 295
Temperature (degrees K)
300
FIGURE 6-lie. Idealized afternoon tenperatnre profile showing
a super adiabatic layer near the surface and a neutral layer aloft.
779
-------
METSCALARS
Of course, available data will seldom be as easy to interpret as the idealized cases
presented above. In addition, data characterizing the vertical structure of the atmo-
sphere is usually available only twice daily. Spatial variations in the mixing height
will also complicate matters. The user must therefore approximate these
parameters from relatively limited data.
In some recent modeling of the Atlanta, Georgia area (Morris et al., 1990a) the diffu-
sion break and temperature gradients were derived solely from standard surface data
and twice-daily soundings. Soundings were available at 0700 and 1900 LSI in this
area. To calculate the temperature gradients, each day was split into two regimes:
a dayiignt regime, during wnich the diffusion break represents the convective mixed
layer, and a nighttime regime, wnen the diffusion break represents the top of the
surface based inversion.
For the daytime regime TGRADBELOW was based on the afternoon soundings for the
days to be modeled and the exposure class. From the soundings the temperature
gradient near the surface was found to range from -.0096 to -.0107 K/m. Based on
tnis, TGRADBELOW was set to -J105 for exposure classes 2 and 3. Since an expo-
sure class of 0 indicates neutral stability, TGRADBELOW was set to -.0098 in this
case. For an exposure class of 1, an intermediate value of -.0100 was used. Since
hourly diffusion break values nad already been estimated, the temperature at the dif-
fusion break (Tm^x) could be calculated using a local surface temperature measure-
ment and the value of TGRADBELOW. From the afternoon sounding the tempera-
ture at the height of the top of the modeling region was estimated and was assumed
to be constant for each day. Then the value of TGRADABOVE could be calculated as
TGRADABOVE = (Ttop - Tmix)/(REGIONTOP - DIFFBREAK)
where
Tmix = Tsurf + DIFFBREAK*TGRADBELOW
^"surf = tne sur^ace temperature
Tto_ = the temperature at the top of the region
90008 19
-------
During the nighttime regime the diffusion break was relatively constant from 200 to
250 meters. It was assumed that the temperature at the diffusion break would
remain constant during the night. Therefore, the temperature at the diffusion break
(Tmjx) was taken from the 250 meter height of the morning sounding. Using a local
surface temperature measurement (TsurA TGRADBELOW was calculated as
TGRADBELOW = (Tmix - Tsurf)/DIFFBREAK.
As an additional constraint, TGRADBELOW was not allowed to exceed the value
found in the first 250 meters (the height of the diffusion break) of the morning
sounding. For this regime, TGRADABOVE was taken directly from the morning
sounding as the temperature gradient from the diffusion break to the region :op. Ail
the above calculations were done using the sounding taken on the morning following
the hour in question, not the one preceding.
6.3.4 METSCALARS Preprocessor (MET5CL)
The overall information flow lor METSCL is shown in Figue 6-12. The MET5CL
preprocessing program requires subroutines from the libraries UTILITY and FILUTIL
(see Section 1.3.1). The program reads a list of file names to be used from unit 5
("standard input") and writes some diagnostic messages to unit 6 ("standard
output"). Use of these two units is confined to the main routine. The list of files
read from unit 5 by METSCL contains just three lines, each with a single file name
and some comment text on each line (Table 6-11). All three lines must be present,
and all three files will be used.
The input data for the METSCL program on unit 3 must be in the standardized for-
mat described in Chapter 4. This file must begin with a CONTROL packet followed
by a REGION packet. A UNITS packet may follow. Within the first TIME
INTERVAL packet, the user must include a SCALARS packet that defines all six
variables.
90008 19
-------
'List of input
and output
file names
Input Data File
CONTROL
/END
REGION
END
(+ other packets)
f METSCALARS [
V (binary tile) V
FIGURE 6-12. Information flow for creating the METSCALARS file.
235
-------
TABLE 6-11. Formac of the METSCL control file (unit 5)
Line
Number Columns
1 Blank or
2 Blank or
3 Blank or
(1-20)
comment
comment
comment
Name
Name
Name
of
of
of
Columns (21-100)
input file
formatted output file
output METSCALARS file
FORTRAN
Format
20X,
20X,
20X,
A80
A80
A80
90003 2$
236
-------
Since the METSCALARS variables do not vary spatially, there is no need to define a
method for determining spatial interpolation. It is assumed that the variables will be
explicitly entered in the 5CALARS packets and that only simple unit conversions,
which are defined in the UNITS packet, need be performed.
Five of the METSCALARS variables, TGRADBELOW, TGRADABOVE, EXPCLASS,
CONCWATER, and ATMOSPRESS, are constant and valid for the entire time inter-
val. The sixth variable, RADFACTOR, however, applies only at the end of the time
interval. Because the UAM calculates radiation factor values continuously over time
by linear interpolation, it also requires values at the beginning of the time interval.
These are saved by the program as the values read for the end of the previous time
interval. Thus the first time interval on the file must end at or before the beginning
simulation time. For axampie, if a simulation is to be made from 0500-1700, and
radiation factor values are input hourly, the values used between 0500 and 0600 are
calculated by interpolating between values input for the 0400-0500 time interval and
those input for the 0500-0600 time interval.
Figure 6-13 shows the packet structure of METSCL. Each of tnese packets .s
described in detail in Section 4.3. Following are soeciai input packet considerations
for the file:
CONTROL
The file name on line 2 must be 'METSCALARS1.
The control variables to be specified on lines 4 to 8 for METSCL are shown in
Table 6-12.
The number of species should be ~ero.
The vertical controls (line 7) should be left blank.
The file unit assignment (line 8) should be left blank. .
The beginning and ending dates and times should reflect the time variation con-
siderations discussed above.
90008 ,. 3
237
-------
CONTROL
END
REGION
END
"UNITS
Optional -
END
TIME INTERVAL
Must appear J \ SCALARS
in the first l' .
time interval
/ ;END
vENDTIME
Can be repeated
nGURE 6-13. Input file structure for preparing the METSCALARS
file.
£E£9CCC8
238
-------
TABLE 6-12. Entries for the
CONTROL packet for the METSCALARS
file.
Line
Number Entry
4 Number of species (= 0)
Spare
Spare
Spare
Spare
Spare
Spare
5 Output file number
Print input values
Print output values
Spare
Spare
Spare
6 Print units cable
Spare
Spare
Spare
Spare
Spare
7 Spare
Spare
Spare
Spare
Spare
Spare
8 Spare
Spare
Spare
Spare
Spare
Spare
-------
A set of output species names is not required; if they are present, their number
must be the same as the entry in the first control parameter on line 4, but they
will be ignored by the program.
REGION, This packet must follow the CONTROL packet. The vertical
parameters will be ignored for the METSCALARS file.
UNITS. This packet, if present, must follow the REGION packet. The UNITS
packet must be provided if any input variable will be input in other than
internal units. The units of RADFACTOR must be min and cannot be
changed.
TIME INTERVAL. Two or more TIME INTERVAL packets must be present.
The first time interval must end at or before the beginning time of any simula-
tion run. All time intervals must be contiguous and of nonzero length. Each
TIME INTERVAL packet contains a SCALARS packet and ends with
ENDTIME. Following the first time interval, only those data that are to be
changed need be specified.
SCALARS. The first time interval must contain a SCALARS packet. Follow-
ing the first time interval, the SCALARS packet can be omitted if no values
are to be changed.
The example input for METSCL shown in Exnibit 6-14 was developed using :ne proce-
aures described. It oegms with CONTROL and REGION packets and includes a
SCALARS packet within each hourly TIME INTERVAL packet. Note that the incut
file covers two full days for the June 3-4, 1984 Atlanta simulation. An extra hour is
included prior to June 3 in case a simulation is to start at 0000'hours on June 3. The
output listing file generated from this input data is shown in Exhibit 6-15.
90008 19
240
-------
IVIL-
a
o
o
o
o
o
o
o
O -OO -00
O ^ OOP^T vft
m v o o i o<*-j
eg o • • —O
•T — O OO
"" 00
.00 -oo
V O O«M ^ ^*
«TOO ' OfM
o • • —o
«T—O OO
"" 00
0)
e
aO OO
CO Z O
o o •
— OO
00 OO
Zwu "OOO
V~ ^* uJt^OW^ —J O taJh-^" Lwi^O*/1-JO l*JH
z ^r vo >—«t—u^ <-» *a X z ^w^>- ctk—v^ujoo za
— Goceoe
t— oe*—I^IUJCD
<0.0
: ^ wi
-mac
<
•H
•^
rH
I
H
OQ
H
241
-------
o
o
4-1
•^oo « co . . «o
T^-O 00
—o
OO
^-o OO
OO
• Oui -o —
^OT O —
(M • • —O
iT» — O OO
oo
*r
i—i
I
0
cct/toc o > u^in oc uo
ujujO VI .J O uJ^— *« ujw
o> UJID oei/toe o> ujirt oet/io:
*) -JO t*j»— ^- tu uj 0 1^1 .^ O »^*— « ujuj O<-
-JO uah-
oci/^oe O > ujt
— t_n*j O <-r> — J O uj^— ^
o >
J O
Lkj _i i_) O u- i
n^— oe>— I^IMJC
JOU-U« <+t —
242
-------
l\J
o
o
o
o
o
o
o
o
o
o
I
I
I
ooui o —
r - • —o
— o oo
• O ^O -O i*1
*o o o rvj tr» en
03 0*0 oo
^ . . t—o
ut —o oo
oo
^- ^- OO
*OCD
oiui
OO
~-O
OO
OO
-OOD -O^-
OO
00
i/»O«— m rfi (vi
oo Lhjtrt oCdoa
VI —JO UJh— i— LU LU
V>< _1QO
1 O U- UJ < <
UJ _JLJ O U.
- —ooae««a.u>-
UJ _JI_IOU_LJ<<
^z «r z loo-ccaeoc
243
-------
o
a
o
o
tx
o
o
(VJ
c
•H
4J
C
.Q
•oo -o^
^ O 2 CM ""^ —
o • . —o
J-> — O OO
00
o
o
~ CO • • -«O
u-v —O OO
O -OO -OCO
O rx. O rx. »-. O r*.
rv rxOOj O rn
— o • • —o
m*-*o oo
oo
o
o
r >oo
n^-oo
o
o
QCunoc o> bum ceuno
uji—tOvi—JO UJK. ^^ wi*rfC
i— oei—onu-ao xzwi^-ceh
JO
rn
wtT) QCU-IOC
O>
> .J O
o -oo • —o
o o1* o o evi ^«. **•>
O COOO ' OCSj
CM — . . O 3
rx —O OO
OO
Lw«r> oe v
I
VD
CQ
M
S
H
<~JOu_(_J<< 1^1
zxoQ-oeoeaox
Ok— «X'OOZZ —
'—' ^ Q£ ^M ^— >— v*j LbJ >—•
244
-------
O
O
1
.g
•o o • — o
*— o oo
oo
1 —J
<
>
w> ^uj oCin
O -OO -OO
i*>tn a£ix>a£ ^o
J *JJ O l^i —i O L*JK- — UJUdO(/1—t
uj os \
o > i^ti
uj —) tj O u- (_> ^
-------
o
o
o
o
4J
C
8
246
3
i
C-. O O
00 •«
vO — o
-OO
Otn
i O *-»
mo
OO
oo
•OO
TOO
»OO
O
O
oo
OO
-o
3^r OO
o • * —o
~O OO
" 00
o
o
Cl
— — ooce «c o-t_>
zxoa.a:ceoax
.(_>« UJ — »(_»Ou_LJ<<
a.ce acoo i < z x oa. oc ce oo
. X(J L3 Z Z — <-JOI— < XO O Z Z
- __
z < sc roo. oc ceoo z C C3 O Z Z — !_J O t— < ->< t3 ^3 z Z — (_> O — < X- O O Z z —
CQ
H
-------
• O'-n
1 O O <
io-n
OO
oo
O -OO -O CM
O — O O **
O ^ O "-O O (""I
— 10 . . _o
VO —O OO
OO
O — O CM n 01 V
— 03O.O O <-nr»
JO
^ CO
^- < 3 UO < _J Q O
UJ _Jl_> O U. LJ ^ <
< 3 V) < — ! Cl C2
CD 3 ^s ft k.1 —
— O
OO
? on 3 i*j oe \o
1 QC^ICC O> "^1
r vi t— K t—1/1 wi co zz«rv>
3K<2.l_l
-------
o
o
o
o
o
o
Ol
o
o
o
o
o
5
I
•O* -OT
ffl O O — O OO
00
o -oo -
O i^or^ —
O • O«T .OO1
o <*noo*~*Oc8
CO >COO OOg
— 00 • • ~-~*
U1 —O OO
O -OO - uji/^ oe^ic
BUJ O£vO
— — eo«
k—IO
1 uj co
248
-------
o
o
51
1
T
I—I
I
• 00 • —O
*J"»OO I O*.
— . . .—o
co~o oo
o
o
« COOO I —r~
CM ~ • • c^o
r^^MO OO
" 00
oa
H
w^uOvO'^O uJ^^rt UJUJO(/1^O
x z wi *- oe t— t/t uj GO
-w* UJ wJ O ft -J O
e wih-oet— i/iujco
coa£
-------
a
X
a,
_)
o
0)
I
-no
O
O
00
Ol
OC CO
0.
bJ O f- 1^
no
>oo
H
CQ
o — w
OOOT
O •
OO
oo
o — o
u^ OC
250
-------
s
5.
0
ce
o
o
LJ
X
z
o
ce.
O
X
-
*
0
ce
0
a
LJ
Z
O
ae
o
X
ae
03
X
Z
l*J
o
z
L3
ac
o
X
<
z
0
ae
o
o
LJ
rf
X
ce
0
z-
0
a
<
z
0
ae
o
0
LJ
^
Z
ae
o
z
o
(J
•^
ae
o
X
—
X
1—
*
—I
-J
_J
—
o
ae
o
>1
z
«
X
h-
0
z
_J
-J
-J
o
•J
ae
a
x
z
^
^
LJ
U,
O
ae
CO
X
— *
o
_J
ac
o
*"
z
^
_J
•^
LJ
O
ae
CO
2;
O
,_J
ae
•NJ
z
^
_j
LJ
Lw
0
ae
UJ
CO
X
^
ae
i
^
oe
0
—i
z
^
_J
•^
u_
o
oe
CO
X
^1
dC
OS
^
zw>»— ae^-*/iu^ea
»eooe— , --
OOOO — OO wOOO — O
ooooooo oooooo
o o
^* *™ .
rf i
2S «^Q£ o» 25 a:^* S»
IZ-rull^alSSil'ffl xz^m-^«2U?^O>-
-------
o
o
OO OOCDCSJO
oooo—«o
ooooooo
oooocomo
oooo wtno
0000—^0
ooooooo
ooooooo
"S
Si
LO
•H
I
-
OX
-------
6.4 SURFACE TEMPERATURE FILE (TEMPERATUR)
The TEMPERATUR file contains time-varying matrices of temperatures at ground
levei. If it is omitted from a simulation, temperature-dependent chemistry calcula-
tions will not be performed (see Section 5.1.2); however, some mass balance and
stability calculations will use a default temperature of 298 K.
6.4.1 TEMPERATUR Preprocessor (TMPRTR)
The TEMPERATUR file is created by the TMPRTR preprocessing program (Figure
6-14). TMPRTR requires subroutines from the libraries UTILITY and FILUTIL (see
Section 1.3.1). The TMPRTR program reads a list of file names from unit 5
("standard input") and writes some diagnostic messages to unit 6 ("standard
output"). Use of these two units is confined to the main routine.
The output variable for the TEMPERATUR file is also named TEMPERATUR. This is
an "implicit output variable" and need not be referred to anywnere in the CONTROL
oacket.
6.4.2 TMPRTR Input Formats
The list of files read from unit 5 by TMPRTR contains just three lines, each with a
single file name and some comment text on each line (Table 6-13). All three lines
must be present, and all three files will be used.
The input data for the TMPRTR program on unit 3 must be in the standardized for-
mat described in Chapter 4. This file must begin with a CONTROL packet followed
by a REGION packet. UNITS and STATIONS packets may follow. Within the first
TIME INTERVAL packet the user must specify a method for each variable and supply
the data necessary for the program to construct the file using those methods. The
methods that can be used to generate the TEMPERATUR file are
90008 .3
253
-------
List of input
and output
file names
Diagnostic
I messages
Input Data File
CONTROL
•
•
END
REGION
END
(+ other packets)
TEMPERATUR {
(binary rile,) V
FIGURE 6-14. Information flow for creating the TEMPERATUR file.
EEE90008
254
-------
TABLE 6-13. Format of the TMPRTR control file (unit 5).
LineFORTRAN
Number Columns (1-20) Columns (21-100) Format
1 Blank or comment Name of input file 20X, A80
2 Blank or comment Name of formatted output file 20X, A80
3 Blank or comment Name of output TEMPERATUR file 20X, A80
3o o o a :5
255
-------
CONSTANT
GRID VALUE
STATINTERP
POISSON
E-WINTERP
N-SINTERP
USER
These methods are discussed in detail in Section 4.1.6. The method most commonly
used for the TEMPERATUR file is STATINTERP.
The time span of the TEMPERATURE file must include the entire time span of the
simulation for which it is to be used. Ground-level temperatures are considered to
De constant during each time interval.
Figure 6-15 shows the input packet structure for TMPRTR. Each of these packets is
described in detail in Section 4.3. Following are special input packet considerations
for the TEMPERATUR file:
CONTROL
The file name on line 2 must be TEMPERATUR1.
The control variables to be specified on lines 4 to 3 are shown in Table 6-14.
The number of species snouid be zero.
If there are input variables that do not appear as output variables, their number
must appear as the number of user-defined variables. All such variable must
also be named in the UNITS packet.
If data from measuring stations are :o be used (methods STATINTERP or
POISSON), the maximum number of such stations must be given.
The number of subregions must be at least one.
The maximum number of parameters must be sufficient to include all specifi-
cations of all parameters.
The vertical controls (line 7) should be left blank.
The file unit assignments (line 8) should be left blank.
90008 19
256
-------
CONTROL
END
REGION
END
f UNITS
Optional
LEND
f STATIONS
Optional < «
LEND ........
; TIME INTERVAL
; SUBREGION
•
*
Must appear ,
in the first «< ; END
time interval i ' METHOD
^
END
'. CONSTANTS
(
Include those packets j • GRID VALUES
appropriate for the j ' •
method(s) selected.
At least one must
appear.
-END
'STATION READINGS
-END
V.' END TIME
Can be repeated
FIGURE 6-15. Input file structure for preparing the TEMPERATUR
file.
257
-------
TABLE 6-14. Entries for the CONTROL packet
for the TEMPERATUR file.
Line
Mumber
Entry
7
I
Number of species (= 0)
Number of user-defined variables
Number of stacions
Number of subregions
Number of parameters
Spare
Output file numoer
Print input values
Print output grid
Spare
Spare
Spare
Print units caole
Print station locations cable
Print regional grid
Print metnods table
Print station values table
Spare
Spare
Spare
Spare
Spare
Spare
Spare
Spare
Spare
Spare
Spare
Spare
Spare
90008 25
258
-------
The beginning and ending dates and times should reflect the time variation con-
siderations discussed above.
A set of output species names is not required; if they are present, their number
must be the same as the entry in the first control parameter on line 4, but tney
will be ignored by the program.
REGION. This packet must follow the CONTROL packet. The vertical
parameters will be ignored for the TEMPERATUR file.
UNITS. This packet, if present, must follow the REGION packet. The UNITS
packet must be provided if:
Any input variable will be input in other than internal units.
Any user-defined variables are specified.
COORD or HEIGHT unit conversions are to be used.
The number of user-defined variables must not exceed the maximum specified
in the CONTROL packet.
The internal unit for TEMPERATUR is degrees kelvin. If the input values for
this variable are to oe in any other units, TEMPERATUR must be speciiiea in
the UNITS packet.
STATIONS. This packet is required if either of the methods STATINTERP or
POISSON is specified. The number of stations listed must not exceed the
maximum specified in the CONTROL packet.
TIME INTERVAL. One or more TIME INTERVAL packets must be present. The
first time interval must begin at or before the beginning of the time span
specified on line 10 of the CONTROL packet. All time intervals must be con-
tiguous and of nonzero length. Each TIME INTERVAL packet contains one or
more of the following packets and ends with ENDTIME. Following the first
time interval, only those data that are to be changed need be specified.
SUBREGION. The first time interval must contain a SUBREGION packet; the
inclusion of this packet in other time intervals is optional. The number of sub-
regions must not exceed the maximum specified in the CONTROL packet.
METHOD. A method must be provided for every variable—including user-
defined variables—in every subregion in the first time interval. Methods can be
changed in subsequent TIME INTERVAL packets if desired. Note that each
parameter must not exceed the maximum specified in the CONTROL packet.
30008 .3
259
-------
CONSTANTS. If the method CONSTANT is assigned to any variable in the
METHOD packet, the first time interval must contain a CONSTANTS packet.
More than one CONSTANTS packet can appear in any time interval.
GRID VALUES. If the method GRID VALUE is assigned to any variable in the
METHOD packet, the first time interval must contain a GRID VALUES
packet. More than one GRID VALUES packet can appear in any time interval.
STATION READINGS. If either the POISSON or STATINTERP method is
assigned to any variable in the METHOD packet, the first time interval must
contain a STATION READINGS packet. More than one STATION READINGS
packet can appear in any time Interval.
In the example input to TMPRTR in Exhibit 6-16, a STATIONS packet is included
after the CONTROL and REGION packets since measured values will be interpolated
to generate the gridded temperatures. Within the first TIME INTERVAL packet a
single subregion is defined and the method STATINTERP is specified for
TEMPERATUR in that subregion. The parameters supplied in the METHOD packet
will ensure that the radius of influence used in the interpolation is large enough to
include ail stations. The STATION READINGS packet follows, with the temperatures
in degrees Kelvin. TIME INTERVAL packets follow at hourly intervals. The only
pacxet needed witnm each succeeding TIME INTERVAL packet for this example is a
STATION READINGS packet to update the temperature values. The input data
covers two full aavs.
6.4.3 TMPRTR Output
The formatted output listing from the example input file is shown in Exhibit 6-17.
Note that since most print options are turned off in the CONTROL packet, format-
ted output from TMPRTR includes, for the most part, only a reiteration of the input
values. The output data file on unit 20 (not shown) will include the gridded tempera-
ture values.
30003 13
260
-------
- 9
o
o
o
o
i^ (MOO O
o-i <\i
1*1
— O
a
o
o
o
o
O
O
o
,00
• >ooo
OOOO
oooo
0*^*0 **>
O 3
o *—
o <
_ v» oc a: ota:
a. ... z 0.0.0.0.
x ooo — xxxx
O ce ac atae
O 3 a 3 3
z a. a. a. a.
— zxxz
wn ce on sc a
z a. a. a_
_ —ZZ XX
3 uj^w -juj < QI-J
ae 2000
3-3 CM
M
ffl
M
one —
ae w e
t— a. ^
— oo
o oo
O — cotfl Of~—oo
4>*JV1(->Q.QXuJH- (yO
-------
3
3
-2
o
a
^i — O O O
-n
T CD ii
« f%l
0-t rv mOO
in o^** co o
u*) CO '•"•- CO O
4J
c
8
o
o
O CCCHCe.CH
O 33 = 3
o oe
O =
O czccac.cc.
O 3 3 O 3
CO *~^-*-^-
vi aeaeoc ae
as o_a_ a, a.
za. 0.0.0.
— ZZXX
aco. 0.0.0.
—XXX X
— —
w^ac cc ex ae
z o. a. a. o.
-I —XXXX
CQ
LU Z T Z
-ODO
J ^
c < « < CD aee
t— CDH-OO -
C uju-11— uo L
t-cQ(—OO3
t/1 <_J Q. O X *-
h-oo^-oozz^ ^-MH-oazz^ —eo*—oo
^1 U ^ Q X —|- bui H- (/I LJ d. O X t*- ^™< *•• Wl O a. O X
-------
o
o
— >j O "" C
T o c a*
M ~i m «NJ
^ -*.~- —XXXX
i^aeaeaeae on at ct ae at
z a_ a. a. a. za.Q-Q.Q-
— XXXX -J —XXXX -J
O '.IL' '-i-f '-u1 ''-' tf O ,^ i,^, ^. i_( ^
h- —
jzrr z
E—000
CQ
H
— •— ffl«— O O Z Z — t—aO»—OOZZ —
263
-------
o
o
o
O
o
§
T OOO •
co »n n f*t
— O
T O
CO <^
-1 OO O O O
— >3 o z 3: — —
264
-------
^O rn ^ OO
O
O
^O -Oi^ O O
O ee.ce.ct.oc.
O =33=3
oe oc oe ^ocaeocoe
0.0.0. z a. a. a. a.
ZXZ -J — ZXZX
h-^-l— > ec oe a a:
z a. o. a. a.
•^
1
•H
4J
s
CQ
H
<— COt— OO
«— CQt— OOZZ— t— O3k-OOZZ— K-O3»— O<3
— *— 33 t— O O
265
-------
O 'D O
O TO^COO
^O 43TTO
OO <•*-» '•n e
^ _r^ ^_O
T OOOl
CD t*i n rsj
'« -^ — <^J O
-n ....
^r OOO i
^O ^-J «T J( O
O ee.cc.ee.ee
O 2=133
O cf.ee.ec.ee.
O 3333
O ccee.of.ce,
O 3333
O oe oe oc oe
O 3333
O ccacccce.
O 3333
o sc.stcc.ee.
O 3333
<-oocae ce ae
zc. 0.0.0.
vnoe oc se oc
z a. a. o. a.
— ZXXX
O^J^juJu^
<)—H-^-t—
O *jj ^ 1^1 VL.
z o. o. o. o.
-J *-XXXX
> cCt->->-h-
(/i ae sc ac cc.
x a. a. o. a.
-QOO X — ODO X —GOO
c «
-------
o --. — ^-®
<0 ^ WOO
T 3 OO
^O <^O ^ O
^O e\j O '-i. O
o •
T o '
GO ?M <*1
O —OOO
— oozz— *— co^-ooaez— f-co^oozz— ^-CDH-
— H-CDK-OOZZ— h-^^-
267
-------
a
o
oa
•a: oa z oc
— oo z a
a. o z bhju
268
-------
-3
O
O
O
O
.u _ — a.
CO <
<7t II i/l Q£
— Z <
w* O a.
• Z —
7 O O O
f> h- CE Z
< CO P—
c ifl in X
2
~) u. u. u.
_l u O
a: o « A3
a. z ^
>- * >-
M . O O
ca _j z >— x
— o
I/I «0
wn *-
uj c
O <0
O —
cc *•*
0. <
a.
*-
a
O
X
ae ce <
h»l uj CC
CO
- O
O Z
— Z
• -OOP
OOO«T
O •
OO
O •
oo
OO
r-
I—(
I
EH
M
m
269
-------
ooooo
ooooo
o
ae
O
19
ae
o
1
•H
4J
0
r~
•H
I
s
270
-------
O
O
o
in OOOOO
ui OOOOO
— OOOOO
*r ivjoooo
OOOOO
OOOOO
OOOOO
O ^O OOOOO r**» OOOOO
O ^> OOOOO k/* OOOOO
O — OOOOO — OOOOO
O V — 0*£>OO ^ VOOOO
I
.H
B
O sccceroe
O 3=300
VI QC S£. QC QC
L3 LU UJ t«J LU
Z Q.Q.O. O.
^ bkrf UJ h*J W ^
z a. a. a. a.
—XXII
: — i—CD^-OOZZ— h-co^
<«cozoox
h-CDK-OOZZ— h— CD*— O O a
CO
H
T*
53
w
271
-------
W V I I
6.5 WIND FIELDS FILE (WIND)
The WIND file contains hourly, gridded horizontal wind components for each of the
UAM layers and scalars representing overall maxima and average values along the
boundaries. Wind fields for the WIND file can be calculated by the Diagnostic Wind
Model (DWM), supplied with the UAM modeling system software, followed by inter-
polation of the DWM wind fields from the DWM layers to the UAM layers with the
program UAMWND. Complete details on the DWM are contained in Volume III of the
UAM guide. This section provides an overview of the DWM and details on the
UAMWND conversion program.
6.5.1 Overview of the Diagnostic Wind Model
The DWM is used to generate gridded fields of the horizontal wind components, u and
v, at several user-specified vertical levels at a specified time. This model
incorporates observations, where they are available, and provides some information
on terrain-induced air flows in regions where iocai observations are absent. The
model is formulated in terrain-parallel coordinates. Wind fields are generated using
a two-step procedure. This procedure is presented in Figure 6-16.
In step 1 a domain-scale mean wind is adjusted for terrain effects. These include the
kinematic effects of terrain (the lifting and acceleration of the airflow over terrain
obstacles), thermodynamicaily generated slope flows, and blocking effects. Step 1
produces a spatially varying gridded field of u and v for each vertical level within the
modeling domain.
In step 2 observational information is added to the (u,v) field calculated in step 1
using an objective analysis procedure: observations are used within a user-specified
radius of influence while the step 1 (u,v) field is used in subregions where observa-
tions are unavailable. This objective analysis procedure consists of the following
four steps: (1) interpolation, (2) smoothing of the analyzed wind field, (3) computa-
tion of the vertical velocity, and (4) minimization of the three dimensional diver-
gence. The following modified inverse-distance-squared weighting scheme (Ross and
Smith. 1986) is used for the interoolation of data:
90003 19
273
-------
Domain-mean
wind
Surface and j_
upper-air data
Diagnostic Wind Model
Step 1
Parameterization of
terrain effects
(kinematic effects,
blocking, slope flows)
Step 2
: Objective analysis
^ (observational information
> is added to the
- Terrain-adiusted flow deid
Minimization of
the divergence
Hourly, gridded
wind fields
FIGURE 6-16. Flow diagram for the Diagnostic Wind Model.
EEE90008
274
-------
E rk-2 + R-2) (1)
where (UO,VQ) denotes an observed wind at station k, r^ is the distance from station k
to a given grid point, (u,v)j is the step 1 wind field at the grid point, and (u,v)' is the
updated wind vector. The parameter R controls the relative influence of the obser-
vations and the step 1 wind field.
Following the interpolation, a five-point smoother is applied to the analyzed wind
field to reduce the discontinuities that may result from the interpolation. An initial
vertical velocity, W1, is calculated from (u,v)' by integrating the incompressible con-
servation-of mass equation. Vertical velocities obtained from an objective analyzed
field may be unreaiistically large near the top of the domain (Godden and Lurmann,
1 983). In the DWM, W1 may be modified using a procedure suggested by O'Brien
(1970):
W2(2) = W(Z) - (Z/Ztop)W Ztop (2)
where Z is the height in terrain-following coordinates and Zr is the height of the
model top. Note that when this procedure is invoked, ^ is zero at the top of the
model. If the vertical-velocity adjustment procedure is not invoked, the final
product of the DWM, (u,v>2, is equal to (u,v)'.
If the vertical profile is adjusted, it is necessary to adjust the objective analysis pro-
duct (u,v)' so that it is mass consistent with W£. An iterative adjustment of the hori-
zontal (u,v) field is performed to minimize the three-dimensional divergence within
each layer. The adjusted horizontal wind field (u,v)7 is the final product of the
DWM.
6.5.2 Mapping of Modeled Wind Fields to UAM Layers (UAMWND)
The preprocessing program UAMWND is a stability-dependent scheme that maps the
modeled wind fields to the layers in the UAM. UAMWND uses the mixing height
30003 . j
275
-------
information provided in the DIFFBREAK file. It is based on two assumptions: (1)
during the daytime vigorous convective overturning within the lowest layers of the
atmosphere produces a well-mixed layer with little wind shear, while during the
evening hours the vertical mixing of momentum is supressed due to the increase in
stability associated with decreasing surface temperatures; and (2) for most applica-
tions of the DWM the surface wind monitoring network is more dense than the upper-
air wind monitoring network.
To take into account vertical mixing and the increased amount of data from surface
monitors, information from the surface-layer DWM winds is incorporated into certain
of the upper layers of the model. The vertical extent of the influence of the sur-
face-layer winds is estimated as a function of wind speed and stability class
az = a(Ut)b (3)
where t is the Lagrangian integral time scale and the coefficients a and b are related
to Pasquiil-Gifford stability class (Panofsky and Dutton, 1984) as indicated in Table
6-15.
Table 6-15.. Relationship between
Pasquili-Gifford stability class and
a and b coefficients.
PG
Stability
Class
A
' B
C
D
E
F
a
0.40
0.33
0.22
0.15
0.06
0.04
b
0.91
0.36
0.80
0.75
0.71
0.69
The Lagrangian integral time scale represents the average lifetime of the turbulent
eddies within me mixed layer. It can be estimated from the relationship
90008 19
276
-------
t =
where ki is approximately 0.15, z^ is the mixing height (as specified in the
DIFFBREAK file), and aw is the standard deviation of the vertical velocity. The
standard deviation a... can be estimated as
Wr
where a. is the standard deviation of the horizontal wind and is related to stability
*
class as indicated in Table 6-16.
Table 6-16. Relationship between
Pasquiil-Gifford stability class
and CK.
PG
Stability
Class a.
A.
3
->
D
£
F
0
0
0
0
0
0
.'74
.122
.087
.035
.017
.009
Substituting the expression aw from Equation 5 into Equation 4 and the resulting
expression for t from Equation 4 into Equation 1 gives the following relationship
between oz, z^, and a^:
(6)
The resulting scale heights for the influence of the surface wind are independent of
wind speed. The scale height, oz, is never allowed to exceed 85 percent of the
mixed-layer height, Zj.
277
-------
The surface-layer wind field is then incorporated into the upper layers according to:
U'k = (1 - wgt(z))Uk + wgt(z)Us (7)
where Us is the surface wind, U^ is the DWM kth layer wind, and z is the elevation of
the center of the kth layer. The weights are estimated using the following set of
relations
wgt(z)= 1.0, .for 0
-------
* W I I 1
velocity adjustment procedure (O'Brien, 1970) is applied, and the horizontal wind
fields are iteratively adjusted to minimize the three-dimensional divergence within
each layer.
6.5.3 UAMWND Input Format
Following the exercise of the DWM, the wind fields are interpolated to the UAM
layers using the UAMWND program. UAMWND requires three input files: (1) the
diffusion break (DIFFBREAK) file prepared for the UAM simulation, (2) the DWM
wind output file, and (3) an input parameter file.
The controlling parameters for the UAMWND interpolation program are specified in
the input parameter file. The parameters are listed and described in Table 6-18; an
example of this file is given in Exhibit 6-18. The information flow diagram for
UAMWND is presented in Figure 6-17. Some guidelines for the specification of the
parameters are given here.
Smoothing is used to reduce the discontinuties that result from the interpolation.
Typically, 2-4 smootning passes (NSMTH) are sufficient to reduce the discontinuities
but preserve the airflow features.
Stability classifications are specified by the parameter ISTAB for each hour. The
exposure class parameter (EXPCLASS in the METSCALARS file) provides a basis for
determining PG stability class (Turner, 1967).
The rougnness length, ZO, is a measure of domain-scale surface roughness. Typical
roughness lengths include 0.2 m for grassland, 0.6 m for suburban housing, and 1-5 m
for forests and cities. (Spatial variation in roughness length is specified in the UAM
input file TERRAIN; see Section 7.1).
Upper limits for the DWM and UAM grid dimensions are specified in the model
through the use of parameter statements (Table 6-19).
90008 19 279
-------
TABLE 6-18. UAMWND input controlling parameters. (See Exhibit 6-18)
Record Parameter
Description
FORTRAN
Format
1 TITLE
2 TOP
3 NZD
4 HGT
NSMTH
IDATEW
3EGTIW
Descriptive identifier.
Top of the UAM modeling domain
(m) as specified in the REGIONTOP
file.
Number of vertical layers in the
DWM wind field.
Heights of DWM vertical layer
interfaces (m) in terrain-following
coordinates beginning with 0 for
the surface.
Number of smoothing passes.
Beginning Julian date of the WIND •
file (yyddd'. where yy is the last
two digits of the year and dcta is
che Julian date).
Beginning cime (specified as hour
on the 2400-hour elocx:),
(60A1)
(10X,F6.0)
(10X,I5)
(10X.10F6.0)
(10X,I5)
POX,15)
POX,F6.0)
8
9
10
11
JDATEW
ENDTIW
ISTAB
ZO
Ending date.
Ending time.
Stability class for each hour of
the interpolation. Numbers 1-6
correspond to Pasquill-Gifford
stability classes A-F.
Domain-scale roughness length (m) .
(10X,I5)
(10X,F6.0)
(10XJOI5)
(10X,F5.2)
continued
90008 23
280
-------
TABLE 6-18. Concluded.
FORTRAN
Record Parameter Description Format
12 NXT Index of x-direction grid cell (10X,I5)
for printing diagnostics.
13 MYT Index of y-direction grid cell (10X,I5)
for printing diagnostics.
30003 ^ 281
-------
O
,0
<
z
<
(MMCS1
O
O
a.
ae
Xw — i^ — a
232
-------
OUTBREAK
file
(binary)
(15)
(8)
Wind file created
byDWM
(binary)
(7)
UAMWND
(11)
Diagnostic
output
message file
UAMWIND
input file
(binary)
Input
parameters
(12)
(9)
Wind
:omponents
listing
(ASCII)
FIGURE 6-17. Flow diagram for the UAMWND conversion program.
283
-------
TABLE 6-19- Parameter statements specified within UAMWND,
Parameter
Description
Stored Value
NXMAX
NYMAX
NZMAX
NZDMAX
Maximum number of grid points
in the x-direction
Maximum number of grid -points
in the x-direction
Maximum number of UAM vertical layers
Maximum number of DWM vertical layers
50
50
90008 23
284
-------
6.5.* UAMWND Output
As depicted in Figure 6-17, the UAMWND postprocessor generates a binary output
file which contains hourly,gridded horizontal wind components (km/h) for each UAM
layer and hourly values for the overall maximum wind speed and average wind speed
along the boundaries. This file is ready for input into the UAM. Additional output
files contain formatted horizontal and vertical wind components and diagnostic out-
put from the layer-matching scheme.
9000319
285
-------
7 INITIAL AND BOUNDARY CONDITION FILES
The UAM requires several files for specifying initial concentrations (AIRQUALITY),
concentrations along the four lateral boundaries (BOUNDARY) and aloft
(TOPCONC), and the characteristics of the underlying surface (TERRAIN). These
initial and boundary condition input files are described in this chapter.
7.1 LAND COVER CHARACTERISTICS FILE (TERRAIN)
The TERRAIN file contains the surface roughness lengths and the deposition fac-
tors. These parameters may vary in the x and y dimensions but not temporally. If no
TERRAIN rile is available to the UAM, the model will use the constant region-wide
default values for surface roughness and deposition factor in the 5IMCONTROL
file. Because the UAM employs a terrain-following coordinate system, terrain neight
information is not used by the UAM and is not contained in this file.
7.1.1 TERRAIN Preprocessor (CRETER)
The TERRAIN file is created by the preprocessing program CRETER from a file,
prepared by the user, containing land use category numbers for each grid cell in the
region (Figure 7-i). A standard UAM file is created containing the variables
ROUGHNESS and VEGFACTOR. The values for ROUGHNESS and VEGFACTOR for
each land use category are built into CRETER (Table 7-1).
The method that CRETER follows for determining the surface roughness lengths and
the vegetation factors is to determine the land use distribution and assign roughness
and vegetation factors based on that land use.
30003 20
287
-------
List of input
and output
file names
(5)
1 (Land use
values for each,
grid cell)
(3)
CRETER
Diagnostic
messages
TERRAIN
("binary)
FIGURE 7-1. Information flow for creation of the TERRAIN file.
EEE90008
238
-------
TABLE 7-1. Surface roughness and deposition factors used by
the program CRETER. Information is based on studies by
Argonne National Laboratory.
Category
Number
1
2
3
4
5
6
7
8
9
10
']
Land Use Category
Urban
Agricultural
Range
Deciduous forest
Coniferous forest
including wetland
Mixed forest
Water
Barren land
Nonforest wetlands
Mixed agricultural and
range '
Hocicy (low shrubs)
Surface
Roughness
(meters)
3.00
0.25
0.05
1.00
1.00
1.00
0.0001
0.002
0.15
0.10
0.10
Deposition
Factor
0.2
0.5
0.4
0.4
0.3
0.3
0.03
0.2
0.3
0.5
0.5
90008 22
289
-------
Both the source data for the land use categories and the resulting gridded land use
data can be at a variety of resolutions and levels of differentiation. Resolution of
the source data will typically be about 100 meters when using 1:250000 scale US
Geological Survey maps. Standard USGS maps differentiate only a few categories of
land use (e.g., urban, bodies of water, forested, etc.). However, USGS Land Use and
Land Cover series maps have many more specific categories (e.g., residential, light
industrial, deciduous forest, coniferous forest, forested wetlands). The source of
land use data may or may not contain all of the categories identified in Table 7-1.
Categories, should be assigned so that the roughness and vegetation factors specified
best represent the underlying terrain characteristics.
The resolution at whicn the land use is specified can vary depending on the modeling
needs and the time available to collect data. CRETER calculates only tne dominant
land use in each grid cell of the modeling region, which is adequate for virtually ail
urban applications. If variation within grid ceils is desired, the user will need to
develop a methodology for specifying ROUGHNESS and VEGFACTOR (see Volume I
for the format of TERRAIN). Region-wide values for the TERRAIN parameters are
not adequate. Since water and land have very different rougnness and vegetation
factors, it is important to at least differentiate these categories. In addition.
because the vegetation factors for urban and forested or agricultural land are quite
different, these areas should be delineated wnenever possible.
7.1.2 CRETER Input Format
CRETER takes two input files. The first specifies the names of the input and output
files as well as a file identifier to be written to the TERRAIN file. Following this
line is a standard REGION packet. The format of the file, is described in Table 7-2.
This file is read from unit 5 (standard input on UNIX systems). Exhibit 7-1 shows the
CRETER input file for the Atlanta example application.
The second file contains the land use categories for each cell in the modeling
region. Each land use value is an integer occupying three columns. Each row of the
region occupies a separate line with up to 99 values per line. The first line of the
90008 20
290
-------
^
*~
f-
4.}
2
.M
jC
X
>1
03
03
0)
SH^
.
J^
O
co
CO
03
O
p
co
c
03
O
0
,_!
i
03
•d
03
e-
E
O
CM
n
03
SM
O
hf)
03
i-J
03
CJ
03
01
3
-o
c
-a
Q, .u
v 1 3 o
a 3 "^
-T^ l^U i
Cd
=s
•13 .
•-• i
-J
-
O
CM
03
r— t
•rH
CM
-U
3
a
£
•i-*
0)
c
o
0
-U
i.
£
•
OJ
1
C.
s
3
0
O
•a
c
03
S
03
4_)
,— 4
j_
-Q 03
§£
03
•z. z
03 T3
^^
o
o
I
5.
e
CO
c
03
i— t
TH
M
JJ
2
tTiJ
c
:— ' C C i
1 ._ ft
"jJ ' —3
•
•z
— *
^
2S
2S
'ja
c~*
0)
"°
j_>
CO
3
y
0
CO
C
13
• —4
•03
,M
tu
-a
03
T5
03
CM
03
03
.a
iO
3
o
I
03
•a
03
0)
03
•a
03
03
03
^!
O
cfl
n (in UTM coordinates) used to locate the
region and all other fixed points (stations,
rces, boundary lines). These numbers are not
computation but are compared to the reference
n other files for consistency.
-< 3 O
00 00 O S-
"H C CO O CO
i. -i CM C
O i— i •»-> •**•
03 c -o so
03 T3 — « 03 — *
.C O O CO w
£-> g a. 3 O
co
03
03
03
C
M—4
T3
O
O
C3
1
X
O
0
Cn
o
^—
1
^
c
0
03
O
o
rH
1
X
ate (meters)
2
•rH
•a
M
O
o
a
i
^
o
o
-U
O
OJ
1
^!
^*
c
o
03
CJ
O
i— l
1
>•>
03
C
o
N
S
=
2
o
oo
1
3
0)
c
o
CS1
CO
c
•iH
bO
03
O
00
03
03 C
S O 30
55 *> ••-(
31 O CO
CO 03 CO
C*
3]
03
O
a
co
T3 CJ
I C
X ~H
03 C
O
i.
03
C
o
o
03 03
O
03
00
03 JJ
H *> 03
3 •-<
a. TT i
C 03
03 03 JS
03
CO CO
CO
e -u
co ^d
co c
03 T3
tJ O
O
03 C3
TJ
03
4J
C
O
CJ
03
a;
o -o
c c
O 1 03
T3
•-H -a
M C
00 CO
03
o
-U
vjj
CO
i.
0)
03 JT
SO O
•-i 03
30 O
M £
O O
-a -c
T1 3
00 C
03
E-i C
30
03 iJ
^^ "-^
r__< —»
-. JJ 30 i.
03 •-•
O S- 50
jJ 03
03 O
r/3 O C
3 -U —I
O
.M
OJ
.£
jj
o
03
CJ
c
03
^
03
03
.
0
0
M
0
•
o
T3
••-I
U
O
291
-------
•o
0>
3
C
Jj
c
o
CJ
•
C\J
1
>—
u
33
4J
C
cu
c
g
O
U
"1
U
?
p
3j
s
3
<— H
o
CJ
TJ
C
oj
C!
5
4J
r— (
SM
cu
§
CO
S
-«^
•1)
_jj
-TJ
C
•M
-a
o
0
o
i
>:
o
o
-"•"•«.
O
*"-
1.
r—
%«^>
C
o
•<— 1
-*-3
••"O
0
0
i— 1
1
>:
^^
n
L,
cu
jj
0)
2
•w^
0)
^J
rd
^
•1-1
-a
t.
O
O
CJ
1
^
o
*
o
=
a"
CM
t
^~
^~
v_ ^
C
o
••— 4
4J
-TJ
O
O
i-H
1
>>
tt-t f*
O -U
(30
0} C
^ [jj
cj -a
'D
^ CO
c
CM -^
O -t
>i 71
•<-^ — ^ •
w _C tf)
J-> 4-> *r-«
C!j X
S — ed
0)
aj a. jc
C C CO CO
CO 05 O - -
— • ~t 1) 0)
73 CO (Q iJ 4J
Z N DO)
O •-• -i = £
O 7) r— t N^ •- 's^, /
0}
"O CO CJ r; £
•rH £ 4J JJ
- ai cu to &a
tio n .c c c
4J CU 0)
CU CO r- 1 1
^T f^ ^.4 | |
CM 4J 0 X >,
0 0
o o
Cl4 Ci-
X—S
^- o
O C\J
«— 1
1 «—
x^x XH>'
1) 0)
N ^
•M -M
co co
1 1
X >,
cu
CO
1-1
rH
0)
CJ
—1
CM
O
CM
i. O
QJ
^2 £M
§ ^
c §
CO C
J=
4J CO
CM j-i
O
T3
33 C
S
CO "*
o a;
(U cQ j^
JS CU JJ
c c
73 "-« ••-!
CU
C !0 CO
•M r-H (^
CM ^ — 1
0) (U 0)
a a o
cu
N
.—4
CO
•n
•
0
o
) ^
cu
CO
cu
c
cO
o
• *
o o -
CJ
^ ^ "i
CO CU 3
JD' J3 C
JJ 4J CO
CO CO -M
3 3 jC
Z Z =M
0 0 0
C — « CO CO
••HO ••-* C\J -M CO
«— 1 !
co i co '— co <—
r— I «— ^-( •— —(CM
-H ' r-( ' <—* *~*
CU CU CU
CJ C CJ C O C
o o o
<*H -M CM -M CM .^
O 4J O -U O -U
CJ O O
i. aj M cu i. cj
D 1- (U 1- 'U i-i
«Q M J3 -r^ ^Q — 4
§T3 S TT S TJ
1 31 31
Z X Z >> Z N
•a
cu
3
C
•i-l
4J
C
o
CJ
CM
CM
39
O
O
-~1
292
-------
a
cu
TD
3
f_J
O
C
o
CJ
•
CM
1
^~
a.
co
cu
^
o
CJ
_)
3
^
o
iXi
S3
C
s
|~H
o
CJ
•a
c
03
S
i— t
i.
CU
j3 CO
3 cl
Z 2
CU TD
« «
M
c
o
q> ^
£ 00
JJ CU
<0
C 9)
^*4 JJ
c
o
•rH
jj
•H
C
•H
CU
a
•H
03
•^
S
fl^
f^
(13
3.
u
cu
03
f-^
£M
cu
3
O
C
CO
rH
• /^
x-^» CU
CO O
i.
^ ^M
jj o
1)
2 4j
50
• • 'al
o j: •
0 O — >
co S co
CU CU -O 3 i-
j"> j3 £ CU
JJ "^ -jJ
>, >> to c cu
O 0
0 2 ->' -'
'_i_, r^j
3
C -~> CO 1 O
._i O -^ CM «— -=r
— i C\J 1
co i co •— ^— ' '—
—(<——(«— ro
rH • — ( ^ t,
cu cu cu
o ^ o t, >> s-
cu cu 03 Ck_t ^) r—\ ^)
O Oj O 03 OJ
— t — t U — 1
3 i. 3 - ^
^ 13 _O 03 '-^H ^
S3 SO. '- 3
3 O 3 Q. 3 O
2—1 23 CO -I
"o
cu
J^
01
3
c
cu
03
i — t
i.
CU
a
a
3
5
co
^^
— t
cu
o
{+H
0
4J
-C
00
cu
e '«
3 t.
-i— * ^J
C CU
S£
o
o
ft ,
3
in
i
T—
-=r
•^_^
j^
cu
>•)
03
r-S
;_
CU
a.
a
s
•
-a
cu
co
£~>
-o
ca
jj
CH
cu
^
_
^ !
id
V
cu
£}
^J
CO
s
--
o
' —
i
x-^
on
1
*—
s^x
t.
o
la
—
£
CO
E-
O
jJ
fd
c
•^4
^
cu
C-i
'-y'
ao
O
o
293
-------
q
4-1
e 100
o -^oo
•— o m ^
«z o
co o •
•a— oo
•— o oo
E-*
M
CQ
M
s
w
294
-------
file is for the top (i.e., northernmost) row of the region with the remaining rows fol-
lowing in decreasing order. Left-to-right on each line corresponds to west-to-east in
the modeling region. This file is read from unit 3.
Exhibit 7-2 shows the land use file for the example UAM application to Atlanta. The
urban areas show up as grid cells assigned category 1. Some areas of water (category
7) are distributed through the region, but the vast majority of the region is category
6, "mixed forest". The land use categories for this region were determined from an
examination of both standard USGS 1:250000 maps of the Atlanta area and USGS
Land Use and Land Cover series maps of the area. Although more land use types are
represented on the maps than in the land use table (Table 7-1), the dominant land use
type in each ceil has been chosen from those listed in Table 7-1.
7.1.3 CRETER Output
The TERRAIN file is written to unit 7. Additional diagnostic information is written
to standard output. Exhibit 7-3 shows the TERRAIN file created from the example
application.
295
-------
296
-------
I tKKAIN
u
M-t
o
0)
OOOT
oo
oo
•ooo ^
oo
00
O*T
a
s
w
oo
OO
•
-------
7.2 INITIAL CONCENTRATIONS FILE {AIRQUALITY)
The AIRQUALITY file contains a three-dimensional matrix of concentrations for any
number of chemical species. Its primary purpose is to provide the set of initial con-
ditions for the beginning of the modeling period. It can also be used to prepare a
three-dimensional distribution of measured data in any time interval for use in data
analysis studies.
7.2.1 AIRQUALITY Preprocessor (AIRQUL)
The AIRQUALITY file is created by the AIRQUL preprocessor (Figure 7-2). AIRQUL
requires subroutines from the libraries UTILITY and FILUTIL (see Section 1.3.1).
The AIRQUL program reads a list of file names from unit 5 ("standard input") and
writes some diagnostic messages to unit 6 ("standard output"). Use of these two
units is confined to the program's main routine and the subroutine OPENA. AIRQUL
uses the standardized input formats described in Chanter '+.
7.2.2 AIRQUL Input Format
The list of files read from unit 5 by AIRQUL contains just six lines, each with a
single file name and some comment text (Table 7-3). All six lines must be present; if
a file is not used, a dummy file name must be supplied.
The input data for the AIRQUL program on unit 3 must be In the standardized format
described in Chapter 4. This file must begin with a CONTROL packet followed by a
REGION packet. UNITS and STATIONS packets may follow. Within the first TIME
INTERVAL packet, the user must specify a method and vertical method for each
variable and supply the necessary data for the program to construct the file using
those methods.
The following methods for interpolating ground-level values of variables, specified in
the METHOD oacket (Section -X2.9), can be used to generate the AIRQUALITY file:
3000320
299
-------
UDIFFBREAK T * *
(REGIONTOP b -
List of
input and
output files
Input Data File
CONTROL
•
•
END
REGION
END
(+Other packets)/
f AJRQUALITY I
("binary) \
FIGURE 7-2. Information flow for creation of the
AIRQUALITY file.
300
-------
TABLE 7-3. Format of the AIRQUL control file.
Line Columns
Number 1 -20
1
2
3
4
5
6
Blank
Blank
Blank
Blank
Blank
Blank
or
or
or
or
or
or
comment
comment
comment
comment
comment
comment
Name
Name
Name
Name
Name
Name
of
of
of
of
of
of
Columns
21-100.
input file
formatted output file
DIFF3REAK file
REGIONTOP file
TOPCONC file
output AIRQUALITY file
FORTRAN
Format
20X,
20X,
20X,
20X,
20X,
20X,
A80
A80
A80
A80
A80
A80
301
-------
CONSTANT
GRID VALUE
STATINTERP
PO1SSON
SPLIT/COMB
E-WINTERP
N-SIN.TERP
USER
These methods are discussed in detail in Section 4.1.6. The most commonly used
methods for the AIRQUALITY file are CONSTANT, STATIONTERP, and, for hydro-
carbon species, SPLIT/COMB.
Since data in the AIRQUALITY file vary in the vertical direction, a vertical inter-
polation method must also be specified for each output variable in each subregion.
The vertical interpolation methods that can be used are
CONSTANT
ABSPROFILE
RELPROFILE
ABSPROFRAT
RELPROFRAT
E-WINTERP
N-SINTERP
VERTUSER
These methods are discussed in detail in Section 4.1.7.' The vertical interpolation
method most commonly used for AIRQUALITY is RELPROFRAT.
The concentrations on the AIRQUALITY file are used as initial conditions for the
UAM. The time span of the file must therefore begin at or before the simulation
starting time and must end after the simulation starting time.
The input packet structure for AIRQUL is shown in Figure 7-3. Each of these
packets is described in detail in Section 4.3. Following are special considerations for
the AIRQUALITY file:
302
-------
CONTROL
END
REGION
END
TUNITS
Optional v •
LEND
TSTATION
Optional x ;
LEND
' TIME INTER
f\ SUBREGION
I ;BND
Must appear i ' METHOD
m the first ^ ' ;
time interval j * *
KEND
j • VERTICAL METHOD
VEND
C. CONSTANTS
.END
• GRID VALUES
Include those packets appro-
GND
,END
. VERTICAL PROFILES
'END
'ENDTIME
Can be repeated
FIGURE 7-3. Tfinut file ^tracture for ^reoarina 'he MRCUALTT"
-------
CONTROL
The file name on line 2 must be 'AIRQUALITY'.
The control variables to be specified on lines 4 to 8 for AIRQUL are shown in
Table 7-4.
The number of species must be greater than zero.
If there are input variables that do not appear as output variables, their number
must appear as the number of user-defined variables. All such variables must
also be named.
If data from measuring stations are to be used (methods STATINTERP or
POISSON), the maximum number of such stations must be given.
The number of subregions must be at least one.
The maximum number of parameters must be sufficient to include all specifi-
cations of all parameters.
The vertical controls (line 7) must include maximum vertical parameter and
profile entries as applicable.
The file unit assignment (line 8) must provide entries for the QIFF3REAK,
REGIONTOP. and TCPCONC files if, and only if, these files are required by
the vertical methods selected.
The beginning and ending dates and times should reflect the rime variation con-
siderations discussed aoove.
A set of output species names is required; their number must be the same as
the entry in the first control parameter on line 4. If either the ABSPROFRAT
or RELPROFRAT vertical method is selected for any variable, the output
species names and the order specified here must match the species names on
the TOPCONC file.
REGION. This packet must follow the CONTROL packet. The vertical •
parameters must be provided for the AIRQUALITY file.
UNITS. This packet, if present, must follow the REGION packet. The UNITS
packet must be provided if:
Any input variable will be input in other than internal units.
Any user-defined variables are specified.
COORD or HEIGHT unit conversions are to be used.
90008 20 304
-------
.11 IV* VJAM.I I I
TABLE 7-4. Entries for the CONTROL packet for the
AIRQUALITY file.
Line
Number Entry
4 Number of species
Number of user-defined variables
Number of stations
Number of subregions
Number of parameters
Spare
5 Output file number
Print: input values
Print output gria
Spare
Spare
Spare
•
6 Print units table
Print station locations table
Print regional grid
Print methods table
Print station values "able
Spare
7 Mumoer of vertical parameters
Mumoer of heights in profile
Print vertical methods table
Print vertical profile tables
Spare
Spare
3 DIFFBREAK file unit number
REGIONTOP file number
TOPCONC file number
Spare
Spare
Spare
90003 22
305
-------
The number of user-defined variables must not exceed the maximum specified
in the CONTROL packet.
STATIONS. This packet is required if either of the methods STATINTERP or
POISSON is specified. The number of stations listed must not exceed the
maximum specified in the CONTROL packet.
TIME INTERVAL. One or more TIME INTERVAL packets must be present. The
first time interval must begin at or before the beginning of the time span
specified on Line 10 of the CONTROL packet. All time intervals must be con-
tiguous and of nonzero length. Each TIME INTERVAL packet contains one or
more of the following packets and ends with ENDTIME. Following the first
time interval, only those data that are to be changed need be specified.
SUBREGION. The first time interval must contain a SUBREGION packet: the
inclusion of this packet in other time intervals is optional. The number of SUD-
regions must not exceed the maximum specified in the CONTROL packet.
METHOD. A method must be provided for every variable—including user-
defined variables—in every subregion in the first time interval. Methods can be
changed in subsequent TIME INTERVAL packets if desired. Mote that eacn
parameter entry contributes to the overall parameter count; the total number
of parameters must not exceed the maximum specified in the CONTROL
packet.
VERTICAL METHOD. A vertical method must be provided for every variable—
including user-defined variables—in every subregion in the first time interval.
Vertical methods can be changed in subsequent TIME INTERVAL packets if
desired. Note that each vertical parameter antry contributes to the overall
vertical parameter count; the total mustnot exceed the maximum specified in
the CONTROL packet.
CONSTANTS. If the method CONSTANT is assigned to any variable in the
METHOD packet, the first time interval must contain a CONSTANTS packet.
More than one CONSTANTS packet can appear in any time interval.
GRID VALUES. IF the method GRID VALUE is assigned to any variable in the
METHOD packet, the first time interval must contain a GRID VALUES
packet. More than one GRID VALUES packet can appear in any time interval.
STATION READINGS. If either the POISSON or STAINTERP method is
assigned to any species in the METHOD packet, the first time interval must
contain a STATION READINGS packet. More than one STATION READINGS
packet can appear in any time interval.
90008 20
306
-------
VERTICAL PROFILES. If any of the profile methods are assigned to any
species in the VERTICAL METHOD packet, the first time interval must contain
a VERTICAL PROFILES packet. There must be a vertical profile defined (or
implied using ALL) for every variable in every subregion for which a profile
method was specified. The number of height-value pairs in any single profile
must not exceed the maximum specified in the CONTROL packet. More than
one VERTICAL PROFILES packet can appear in any time interval.
If any vertical method other than CONSTANT is selected, the DIFFBREAK and
REGIONTOP files will be read by AIRQUL. In addition, if either of the vertical
methods ABSPROFRAT or RELPROFRAT is selected, the TOPCONC file must also
be read.
The input file for AIRQUL for the example application is shown in Exhibit 7-4. Sines
the example simulation began at 1200 on June 3, 1984, one time interval from 1200
to 1300 is specified. Following the standard CONTROL and REGION packets, a
UNITS packet specifies that the inputs values for NO2> O-j, and CO will be in parts
per billion (ppb). Otner species must be in the standard units of parts per million
(ppm).
Following the UNITS packet, a STATIONS packet lists eight stations, four of wmcn
are the locations of actual monitoring sites at DeKaib, Columbus Airport, Conyers
Monastery, and Dallas, Georgia. The other four stations, located near the four cor-
ners of the modeling region, are fictitious. Data for these sites are used to bring
interpolated values of pollutant concentrations down near the oackground values out-
side of the urban areas. This is not the only way this could be accomplished using
AIRQUL. Subregions could be defined outside the urban areas and the CONSTANT
method applied to these subregions. The use of fictitious stations with an interpola-
tion method, however, results in a smoother transition from urban to rural concen-
trations.
Within the TIME INTERVAL packet a single subregion is defined covering the entire
region. The method STATINTERP is initially used for interpolation for all species.
Other methods may then replace the STATINTERP method for other species further
down in the input file.
307
-------
tji o» ^- --• o a» -o-»c^«Cu_it-vioo«rvjz zuj .^ooo zzomozi—
CO»-iOQ.)-XU.«_iXOQ.za.Xw-»_»HX ujQCO«0*3''=re^'-J=JZOi->u-vi
_
ooooo
—tOQOQ
|_ = 33=JCDcaQC
-
l uj QC
JO X CO
:— n
H
CQ
-------
MlnUUML.1 I T
— OOOOOOOOOOOOOOOOOOO 3
o ooooooooooooooooooo o
O OOOOOOOOOOOOOOOOOOO O
o ooooooooooooooooooo o
ooo oooo
<
: z z z z z z z z z z z z z z z z —
3
- ^^v^^^^o-^^^^^u^ i 00^.00-0
_l ZZZZZZZZZZZZZZZZZZZ — ^PZSS0.0,"
0. OOQOOOOOOOOOOOOOOOO 1—1 OOOOOOO
OOOO
oo.—^
_
t—_J-O—i
03
H
« -,0-
ae x < z «
I— (/I
j_l_j_l_J_J_*_J_)_J_J_J—1—iZ-J-JZO—'—1—'-
C ««<<<<«<« ^ > ^ uj i_)
LJ ^-r-^^ O
)— t—
j a oe -J a
-------
_t — ac >- -jc
i
3
U
CQ
H
w
310
-------
The vertical method selected is RELPROFRAT. Using this vertical interpolation
method, the surface value, whether a result of interpolation or assignment by
CONSTANT, will be specified for the lower portion of the region and the values in
the TOPCONC file
-------
3OOO O
fl
>— o —
1— GO —
3 CP U_
a. —• uj
— uj O — —
CQ
oc =s c
H-CJ «0
Z QC —
O — *-*
xujae_>_jocQuj_jbu
312
-------
I I
t— t— oc
ea — uj
Q. k- —
Lt_ h— O
O •«
QC O UJ
GJ
tt
313
-------
— 0000
3
•*
o
J
aC
O
w
• o
oe
o
o
X
z
s
o
X
t_
o
0
w
z
o
QC
o
o
~
z
0
oe
o
X
H»
f
CO
X
=3
UJ
O
rsj
Z
O
ae
o
X
o
2
0
.0
£!
z
a
ae
o
o
X
z
—
o
0
o
UJ
ae
o a
o o o
o o o
-TOO
-O O O
-O T T
i-«j Z Z
5 - -
ae ae ae
o — —
o a o
< x >•
•x. — - —
— u- |_
0 Z Z
Z _J -J
o
— _J -J
UJ UJ UJ
? ? n
z z z
o o o
^ U4 -J
ae ae QC
a o o
z z z
VI VI VI
-J — ) -J
o o o
ae ae ae
UJ UJ UJ
CO CO CO
XXX
3 3 =3
z z z
<\J
^
3-
5
—1
ae
je
O
z
vi
UJ
O
UJ
CO
X
o
'^
ae
^
5
_j
2
ae
ae
a.
a.
z
VI
_J
UJ
o
ae
UJ
ca
X
z
3
o
ae
**
-j
UJ
u_
ae
1/1
o
0
3
o
UJ
0=
o
_i
Z
h-
X
O
UJ
X
UJ
=3
X
Z
X
0
o
.n
UJ
^
Q*
a.
^
— •
h-
S
(J
UJ
I
UJ
•J-
rs
X
X
oooo ' *
oooo * *
. . . . * *
oooo * *
- *
OOOO OOOOOOOOO 3 —0 0 0 ^000
gggg oao0000.00. ? ' § o §§S§
OOOO OOOOOOOOO r- 0 0 OOOO
oooo* o o oooo
OOiOO OO OO — • u~* O^^O un ae -*-
oooo oooocno-jao — — a hZ~'T~
oooo ^oooa^oiasca T — — ' ,? »?
OOOO -^ ^ — .— •rncT^'-^ — ~ ~ ^ ~ ^
^^SSSSSS fooo = i-^o§SS
Q^QQQ^^f^.^^ Q OOOO ""I
ggggS^SS - OOgg 0
SooSSJCSr^^ _
cacaeo — ' -1 _ ^""^^
Q-Q-a_ < rf z u^OQ-°
Q.O.CL >^ < ^^ Z UJOQ-^-
ujiri z o o vi
vn »- — 0 — z =>
ooooo — -aots o H- « vi o
(/) W—QOOO uj ozoeo^
•
'S
3
C
c
o
u
*
LO
i
g_i
M
CO
HH
^
pa
314
-------
oooooooaooooooo
OO3OOOOOOOOOOOOO OO
•^ooooooooooooooo oo
O3OOOOOOOOOOOOOO OO
o o o oo oo o oooo o oo o oo
oooooo'oooooooooo oo ^rMl/,^_0
'— OOOOOOOOOOOOOO
oo— ooooooooooooooooo
oooooooooooooooooooo
oooooooooooooooooooo
I
•H
I
oooo
o o o o
oooo
ooo o
LO
I
r-
315
-------
— — o o — —
o — —
0)
3
oooooooooooooooo
OOOOOOOOOGOOOOOO
-nocrooooooooooooo
^-ooooooooooooooo
O
z
X
w
316
-------
Mli iVJUMLI I I
— o
<3
X
t—
<£
0.
0
o
0
CM
in
T
CO
x
£
Q
01
-
-*
"Z
:>
3
CD — •
31 -J
Q.
o -x.
c o
3 —
—1
4- X
*- 0.
2
t— z
5 „
QC Z3
>
I
I >-
UJ —
_J QC
U. «t
VJ O)
^ o
o
o
— O>
t— O
0
o
o
O CO
^- 0
o
0
o
o» oo
o
o
0
o
CO ~
o
o
o
o
o
=
=
o
=>
3
o
3
0
o
o
0
m ^0
o
o
o
0
CM iO
o
o
o
0
o
LlJ
u o
0
oc
o
oto»a»o>ooO'-'-"^(VJrnrn.***T
Ililiilliiiilli
ooooooooooooooo
SSoSSSSSSoooooo
000000000000000
ooooooooooooooo
1 i 1 1 i 1 i i I i i i I i i
ooooooooooooooo
i i i 1 i 1 i i i i i I I i i
o'oo'oo'oaoooooooo
^^.^-»-»3ia>01J100 — — ^4"M
12iii3§oioooooo
Io35o0003000000
ooooooooooooooo
1 1 1 i 1 1 i i 1 i i I I I I
ooooooooooo o ooo
il 3 ooooooooo ooo
^0=17500000003000
000 300000000300
OOOO03O:SOOO~ — — —
ooooooooooooooo
ooooooooooooooo
ooooooooooooooo
0--a-331'31000 —
00000000300 — — ^' —
§3i^3i§§l§ooo31
000000003000000
u^«r«T^r^-O'^^:o<^01000^"
§§§§§§§Si§§S33S
ooooooooooooooo
ooooooooooooooo
o->ir»aikr»^LO'**f^^'^^o<3'-3
000000000000 — — —
i31Iiooi?3§ii3i
003030033000303
§§§§§§§§§§§§Soo
ooooooooooooooo
ooooooooooooooo
o>«i^«oi^i'^(v-' — oencor^^o^
'
0
o
o
o
o
o
0
o
0
o
=
o
0
0
0
o
0
^
3
i
o
—
3
o
o
3
o
=
o
0
o
0
o
cst
o
o
o
o
o
0
0
o
o
o
o
0
o
o
o
0
0
0
3
0
o
0
0
—
0
o
o
0
0
o
0
o
o
o
CM
m
0
o
o
o
o
o
o
0
o
o
o
o
i
o
3
o
o
3
o
3
o
o
—
3
o
0
o
o
3
3
o
o
0
CM
o
o
o
o
o
0
o
o
3
0
o
o
o
3
o
0
o
3
.
0
o
0
t"
=
o
0
o
o
3
o
o
o
o
CM
o
o
o
0
o
o
0
o
o
o
o
0
o
o
o
<3
0
o
^
i
3
-J
^
3
o
o
0
o
o
o
o
o
0
o
0
0
o
o
o
o
o
o
0
0
o
0
o
i
o
0
o
o
3
o
0
o
"T
0
0
0
0
0
3
o
o
o
Q
a\
o
o
o
o
o
o
0
o
o
0
o
o
0
0
o
o
o
3
3
3
3
o
o
0
o
o
o
°
0
o
0
oo
o
o
o
0
o
o
0
o
o
o
o
o
0
0
o
0
0
=
~"
,,J
o
o
^
3
=
0
o
=
°
o
o
o
^
0
o
o
0
o
o
o
0
0
o
0
o
3
0
3
3
o
^
"^
3
-4
3
0
"^
0
0
o
0
0
o
o
3
o
o
o
s
ffl
H
w
317
-------
OL
O
O
ZJ
C
3
i-S
*
£
0*
1
X
z
u-l
0 —
X
ae
ce
a.
X.
VI
u-l
-J UJ
o* a
Ct CE
O
0
0
o
0
0
0
0
o
o
o
0
o
o
o
o
o
0
o
o
0
0
o
o
o
o
0
0
o
0
o
o
ooooooooooooooooo
ooooooooooooooooo
OOOOOOOOOOOOOOOOQ
ooooooooooooooooo
ooooooooooooooooo
ooooooooooooooooo
ooooooooooooooooo
ooooooooooooooooo
ooooooooooooooooo
ooooooooooooooooo
ooooooooooooooooo
ooooooooooooooooo
oooooo ooooooooooo
ooooooo oo OG oooooo
ooooooooooooooooo
ooooooooooooooooo
ooooooooooooooooo
ooooooooooooooooo
ooooooooooooooooo
000 OOOOOOOOOOOOOO
00 00 0000000000000
ooooooooooooooooo
ooooooooooooooooo
ooooooooooooooooo
ooooooooooooooooo
ooooooooooooooooo
ooooooooooooooooo
ooooooooooooooooo
0000 — — — c\jU-»00
o o o o ^oooooooooooo
ooooooooooooooooo
ooooooooooooooooo
ooooooooooooooooo
ooooooooooooooooo
ooooooooooooooooo
0
o
CNJ
o
0
0
0
o
o
m
o
o
o
o
o
o
o
o
0
0
0
o
o
o
o
0
03
o
o
o
o
o
3
a
o
o
o
o
o
o
0
o
o
o
0
o
0
0
-y
0
o
0
o
0
o
o
o
o
o
o
o
o
o
CD
0
o
0
o
o
0
o
o
o
0
0
o
o
0
CO
0
o
0
o
0
0
o
o
o
0
o
o
o
o
o
o
o
31
0
o
CO
o
0
0
o
0
o
0
0
o
Csl
0
o
o
l-l
0
o
o
o
o
o
CO
0
o
og
o
o
0
o
o
o
0
o
o
o
o
-0
o
o
01
o
0
o
o
o
0
o
0
0
a*
o
o
o
("I
0
o
o
o
0
0
0
o
o
CNJ
o
0
o
o
o
o
o
o
-o
o
o
o
o
o
0
o
o
o
CO
0
o
o
o
o
o
CO
o
rn
o
o
o
o
o
0
o
o
o
o
o
o
o
o
0
0
o
o
o
o
o
0
0
o
o
03
o
0
o
0
o
o
o
0
0
i
0
0
o
0
o
o
0
0
o
o
o
0
o
o
o
o
o
o
o
o
o
0
o
0
o
0
o
0
o
o
*
1
S
in
EXHIBIT
318
-------
Mil iV*US-\l_l I I
o
o
o
o
o
o
o
r*k 00
a o
o o
o o
o ~ —
o o o
o
o
o
US
o
o
o
o
o
o ^ —«
f* r-* 00 03 00
O
O
o
o
o
o
o
o
o
o
3
O
3
3
O
O
O —
3 O
O
O
1 3
09
O
01
o
o
-------
o
o
o
000
o o o
O —• —
f^ CO
O 0
o o
o o
o
o
o
o
m
I
P3
H
(_J Q
o
— O
-------
- .1 I I
I I
<
a.
"3
-J\
"O
It
c
3
"•}
<«
0
z
VI
-.
a.
ui
•x.
0
_l
X
oc
LU
VI
ac
•a:
0.
z
1/1
<
>
1 —
_l
3
cr
oe
<
•3
IO
^n
T
t—l
_
_J
_J
UJ
0
C£.
O
Q
3
0
O
O
O
•rr
o
0
0
0
o
o
~>
o
o
CNI
0
o
3
3
o
o
3
o
0
J-
o
0
o
-3
O
o
o
o
o
0
o
o
0
3
o
o
cy
O
0
o
0
o
a
^i
0
3
3
eg
3
o
3
o
o
o
3
3
O
J
o
0
0
3
o
o
o
o
o
o
0
o
o
0
o
1*1
3
o
a
o
o
o
-^1
o
o
3
O
O
O
01
0
o
o
0
-3
•o
a
;3
O
3
3
3
O
3
O
o
o
o
o
o
o
GO
0
o
o
o
o
o
o
o
0
0
0
0
o
0
o
0
0
T>
o
o
o
CD
0
o
0
o
•3
o
3
o
3
O
O
-o
3
O
0
o
0
0
o
o
3
o
0
o
o
o
0
o
o
0
o
0
o
o
o
o
o
o
o
o
ca
0
-D
o
o
o
o
a
3
•3
3
J
3
0
0
^-1
3
3
O
O
O
o
o
0
0
o
Z5
0
o
o
o
0
o
o
o
o
0
0
o
a
o
o
o
0
o
:D
3
O
0
0
o
0
0
3
3
3
-T
O
a
o
T
3
3
-rt
0
O
o
o
-3
3
O
O
-O
0
o
o
o
o
o
0
0
0
o
a
•*
Q
O
O
<*
o
o
o
Q
^0
3
3
3
O
O
o
o
t3
3
^T
O
o
3
3
3
V
0
O
0
0
o
o
o
^o
o
o
0
o
0
o
o
o
0
o
o
01
o
0
o
o
a
o
o
a
3
3
3
O
a
o
3
3
3
O
.n
3
3
3
T
-3
3
«T
O
O
o
o
-3
O
O
O
O
o
o
o
0
0
o
o
o
o
0
0
31
0
o
o
a
a
0
0
en
o
o
0
0
o
0
o
3
3
3
4T
3
O
3
3
3
ul
O
O
0
0
o
•o
3
O
o
o
o
a
o
o
o
o
o
0
o
o
a»
o
0
o
o
0
o
o
:o
0
o
•3
O
0
o
o
-3
3
3
^3
"3
ID
3
O
3
3
<-O
0
o
o
o
o
3
3
O
rx
0
O
o
0
•
c
•d
c
•~\
0
in
I
B
CQ
H
K
?N
w
321
-------
^O O ^
w fl
<
0. m
U">
Z
LbJ
t—
csi
o.
o
VI
O
o
—« tM
in
Ul
•T
CD
<
Q
>—
,/•)
—1
-«J
j:
;>
3
o 3
"^i x.
j-\
03 —
Oi — »
i.
vl
rn
e o -
3 —
"I _(
_J
«3 —
^ X
c
-a 3C
«s
as
«— z
C3* '_ij
QC =3
^- _J
< <
i >-
k—
i^j —
X -J
CT
—i oe
u. <
mo»^on>oa'-fM— .oo^a>f^^ui'Tne\j^
oe
'S
1
'4-1
c
0
a
LTI
1
EXHIBH
322
-------
r». O ^O
s>
X
a T
rn
a.
o
o
o
esj fi
m
CO
a
3;
-^
T -M
03 —
CL
e o r^
a —
—5 —1
«-> X
-^ 3C
a.
< Psi
i >•
0* 0
l*> —
_j ac c*
_ _ UJ
-L. <
o
a
0
o
o
a
0
o
o
o
o
o
33
=
3
O
3
3
3
|
0
-
O
o
o
:NJ
O
m
O
O
0
in
0 O O
o o o
0 O 0
000
0 O 0
O O 0
•~O 01 =0
— O 3
030
000
003
000
0 O 0
0 O O
%i — • 3
333
330
333
O O O
3 O 0
333
333
333
333
330
333
333
333
303
030
300
0 O 0
330
333
000
000
o o o
0 O 0
^» \o in
-
urt
O
O
0
0
o
o
oo
o
0
0
0
0*
0
o
o
0
o
o
o
0
o
o
o
o
o
o
o
o
0
0
o
0
0
o
o
o
o
o
o
o
o
0
o
0
o
o
00
o
0
o
o
o
o
o
o
o
o
o
o
o
0
0
o
o
o
o
0
o
o
o
o
0
0
o
o
o
0
o
0
o
o
o
o
0
CO
o
o
0
o
o
o
0
o
o
o
0
0
o
0
0
o
o
o
o
0
o
o
o
o
o
o
0
0
o
0
o
o
o
o
oo
o
0
o
oo
o
o
a
0
o
0
o
o
o
o
o
3
0
o
o
o
0
o
0
o
o
-
c
Tj
c
o
CJ
un
1
EXHIBIT
323
-------
^f tr m fsi CNj "" O o*
—.^- — — ^-^- — o
OOOQOOOO
OOOOOOOO
O
O
O
EH
H
a
324
-------
7.3 LATERAL BOUNDARY CONDITIONS FILE (BOUNDARY)
The BOUNDARY file contains a time-invariant definition of the boundaries of the
region to be modeled and time-varying matrices of pollutant concentrations in each
boundary cell. These boundary values are used by the UAM to represent pollutant
concentrations passing over the boundary into the modeling region.
7.3.1 BOUNDARY Preprocessor (BNDARY)
The BOUNDARY file is created by the BNDARY preprocessing program (Figure
7-4). It reads a list of file names from unit 5 ("standard input") and writes some
diagnostic messages to unit 6 ("standard output"). Use of these two units is confined
to the main routine and the subroutine OPENS.
7.3.2 BNDARY Input Format
The list of files read from unit 5 contains just six lines, each with a single file name
and some comment text (Table 7-5). All six lines must be present; if a file is not
used, a dummy file name must be supplied.
The input data for the BNDARY program on unit 3 must be in the standardized for-
mat described in Chapter 4. This file must begin with a CONTROL packet followed
by a REGION packet. UNITS and BOUNDARIES packets follow. Within the first
TIME INTERVAL packet the user must specify methods for interpolating ground-
level and aloft values for each variable and supply the necessary data for the pro-
gram to construct the file using those methods.
The following methods for interpolating ground-level values of variables, specified in
the METHOD packet (Section 4.2.9), can be used to generate the BOUNDARY file:
30008 20
325
-------
\DIFFBREAK
(11)
REGIONTOP
List of
input and
output files
Input Data File
CONTROL
•
•
END
REGION
END
(+ other packets) j
AIRQUALITY (
V (binary) V
FIGURE 7-4. Information flow for creation of the BOUNDARY file.
326
-------
t I
TABLE 7-5. Format of the BNDARY control file.
Line
Number
Columns
1-20
Columns
21-100
FORTRAN
Format
1
2
3
4
5
6
Blank or comment
Blank or comment
Blank or comment
Blank or comment
Blank or comment
Blank or comment
Name of input file
Name of formatted output file
Name of DIFFBREAK file
Name of REGIONTOP file
Name of TOPCONC file
Name of output BOUNDARY file
20X, A'80
20X, A80
20X, A80
20X, A80
20X, A80
20X, A80
90008 22
327
-------
CONSTANT
BOUNDVALUE
SPLIT/COMB
E-WINTERP
N-SINTERP
USER
All these methods are discussed in detail in Section 4.1.6. The methods CONSTANT
and BOUNDVALUE are equivalent. The methods most commonly used for the •
BOUNDARY file are BOUNDVALUE and, for hydrocarbons, SPLIT/COMB. The
methods E-WINTERP and N-SINTERP specify a linear interpolation between the
values at the endpoints of the boundary Line and are equivalent interpolation methods
for BOUNDARY.
Since data in the BOUNDARY file vary in the vertical direction, a vertical interpola-
tion method must also be specified for each output variable in each subregion. The
methods that can be used are
CONSTANT
ABSPROFILE
RELPROFILE
ABSPROFRAT
RELPROFRAT
E-WINTERP
N-SINTERP
VERTUSER
These methods are discussed in detail in Section 4.1.7. The methods most commonly
used for the BOUNDARY file are CONSTANT and RELPROFRAT. The methods
E-WINTERP and N-SINTERP specify a linear interpolation between the values at the
endpoints of the boundary line and are equivalent interpolation methods for
BOUNDARY.
The time span of the BOUNDARY file must include the entire time span of the simu-
lation runs for which it is to be used. Boundary concentrations are considered to be
constant during each time interval.
90008 20
328
-------
The input packet structure of the BNDARY preprocessing program is shown in Figure
7-5. Each of these packets is described in detail in Section 4.3. Following are
special input packet considerations for the BOUNDARY file.
CONTROL
The file name on line 2 must be 'BOUNDARY1.
The control variables to be specified on lines ^ to 8 in BNDARY are shown in
Table 7-6.
The number of species must be greater than zero.
If there are input variables that do not appear as output variables, their number
must appear as the numoer of user-defined variables. All such variables must
also be named in the UNITS packet.
The number of boundary line segments must be at least three.
The maximum number of parameters must be sufficient to include ail specifi-
cations of ail oarameters.
The vertical controls (line 7) must include maximum vertical parameter and
profile entries as appiicaole.
The file unit assignment (line 3,) must provide entries for the DIFFBREAK,
REGIONTOP, and TOPCONC riles if, and only if, tnese files are required by
the vertical method's selected.
The beginning and ending dates and times should reflect the time variation con-
siderations discussed above.
A set of output species names is required; their number must be the same as
the entry in the first control parameter on line 4. If eithr the ABSPROFRAT
or RELPROFRAT vertical method is selected for any variable, tne output
species names and the order specified here must match the species names on
the TOPCONC file.
REGION. This packet must follow the CONTROL packet. The vertical
parameters must be provided for the BOUNDARY file.
UNITS
This packet, if present, must follow the REGION packet. The UNITS packet
must be orovided if:
90008 20
329
-------
CONTROL
END
REGION
END
fUNTTS
Optional •s •
LEND
BOUNDARIES
Must appear
in the first
time interval
END
! TIME INTERVAL
METHOD
,END
•VERTICAL METHOD
VEND
f> CONSTANTS
Include those packets appro-
priate for the method(s) selected.
At least one must appear ^
;END
- BOUNDARY READINGS
;END
' VERTICAL PROFILES
;END
VEND TIME
Can be repeated
FIGURE 7-5. Input file structure for preparing the BOUNDARY file.
330
-------
TABLE 7-6. Entries for the CONTROL packet for the
BOUNDARY file.
Line
Number Entry
4 Number of species
Number of user-defined variables
Number of boundary line segments
Spare
Number of parameters
Spare
5 Output file number
Print input values
Print output boundary values
Spare
Spare
Spare
6 Print units table
Print boundary line segment locations table
Print regional grid
Print oiethods table
Print boundary values table
Spare
7 Mumoer of vertical parameters
Mumber of heights in profile
Print vertical methods caole
Print vertical profile cables
Spare
Spare
8 DIFFBREAK file unit number
REGIONTOP file number
TOPCOMC file number
Spare
Spare
Spare
90003 22
331
-------
Any input variable will be input in other than internal units.
Any user-defined variables are specified^
COORD or HEIGHT unit conversions are to be used.
The number of user-defined variables must not exceed the maximum specified
in the CONTROL packet.
BOUNDARIES. This packet is required; it names the line segments that define
the boundaries of the region. The number of line segments specified must
equal the number specified in the CONTROL packet.
TIME INTERVAL. One or more TIME INTERVAL packets must be present. The
first time interval must begin at or before the beginning of the time span
specified on line 10 of the CONTROL packet. Ail time Intervals must be con-
tiguous and of nonzero length. Each TIME INTERVAL packet contains one or
more of the following packets and ends with ENDTIME. Following the first
time interval, only those data that are to be changed need be specified.
METHOD. A method must be provided for every variable—including user-
defined variables—for every boundary line segment in the first time interval.
In the METHOD packet, the boundary line segment name is entered in the "sub-
region" field. Methods can be changed in subsequent TIME INTERVAL packets
if desired. Note that each parameter entry contributes to che overall
parameter count; the total number of parameters must not exceed the maxi-
mum specified in the CONTROL packet.
VERTICAL METHOD. A vertical method must be provided for every variable-
including user-defined variaoies—for every boundary line segment in the first
time interval. In the VERTICAL METHOD packet, the boundary Line segment
name is entered in the "subregion" field. Vertical methods can be changed in
subsequent TIME INTERVAL packets if desired. Note that each vertical
parameter entry contributes to the overall vertical parameter count; the total
must not exceed the maximum specified in the CONTROL packet.
CONSTANTS. If the method CONSTANT is assigned to any variable in the
METHOD packet, the first time interval must contain a CONSTANTS packet.
More than one CONSTANTS packet can appear in any time interval.
BOUNDARY READINGS. If the method BOUNDVALUE is assigned to any
variable in the METHOD packet, the first time interval must contain a
BOUNDARY READINGS packet. More than one BOUNDARY READINGS
packet can appear in any time interval.
90008 20
332
-------
VERTICAL PROFILES. If any of the profile methods are assigned to any
species in the VERTICAL METHOD packet, the first time interval must contain
a VERTICAL PROFILES packet. There must be a vertical profile defined (or
implied using ALL) for every variable for every boundary line segment for
which a profile method was specified. In the VERTICAL PROFILES packet, the
boundary line segment name is entered in the "subregion" field. The number of
height-value pairs in any single profile must not exceed the maximum specified
in the CONTROL packet. More than one VERTICAL PROFILES packet can
appear in any time interval.
If any vertical interpolation method other than CONSTANT is selected, the
DIFFBREAK and REGIONTOP files will be read by BNDARY. In addition, if the
vertical method ABSPROFRAT or RELPROFRAT was selected, the TOPCONC file
will also be read.
The input file for BNDARY for the example application is shown in Exhibit 7-6. In
the example the boundary conditions cover the entire span of the Atlanta simulation
of June 3-4, 1984. The time span of the file and the first and only time interval are
ooth set to begin on January 1. 1980 and end on December 31, 1989 so that it will
more than cover the span of the simulation. The file begins with the standard
CONTROL and REGION packets, followed by a BOUNDARIES packet that consider-
ably reduces the number of grid ceils to be simulated from the maximum ^O-by-^O
cell region. The simulated portion of the region will include the cells designated as
the boundary ceils and the ceils within the domain formed by these boundary ceils.
Since no UNITS packet is included, the input values for ail species will be expressed
in internal units.
The interpolation methods for all variables are BOUNDVALUE for ground-level
values and CONSTANT for values aloft. Although all boundary line segments are set
to the same value in the example, they could be set individually to different values.
With the CONSTANT vertical method, variables in all five layers above ground will
have the same value as at the surface.
90008 20
333
-------
coo-— —oo
O
o
o
o
o
OOOO
0000
oooo
0000
oooo
^ oo o
0000
oooo
0000
; O O T o c
oo o o c
1 _J CM X i-J £
: _j o o ft — —i <
J < O>-O —-c
OCVjfl
zoo
ujZ— UJ— iZ-J— IZO
z;
3
•H
0)
rH
a
i
EH
M
m
334
-------
335
iH
§
EH
H
CQ
-------
7.3.3 BNDARY Output
The output variables for the BOUNDARY file are the chemical species named in the
CONTROL packet. Additional user-defined variables (e.g., "reactive hydrocarbons")
can be specified in the UNITS packet. The internal units for the concentrations of
all species except aerosols (AERO) are parts per million (ppm); for AERO, the units
are micrograms per cubic meter (ug/m ). The standard names for reactive species
recognized by the UAM are listed in Table 5-2. If any of these species does not
appear on the BOUNDARY file, the boundary concentrations used by the UAM will
default to a value defined in the CHEMPARAM file. Any species in BOUNDARY
that do not also appear in CHEMPARAM will be ignored.
A portion of the formatted output from the example input file is shown in Exhibit
7-7. The example was printed with the grid option. The user should exercise caution
when selecting this option, since a very large output file can result. The example
printout contains a gridded map that denotes the boundary cells for the "south",
"east", "north", and "west" boundary line segments. For the user-defined segments in
this example for the Atlanta domain, I has been designated for the south boundary, 2
for the east, 3 for-the north, and 4 for the west.
The exhibit also shows the boundary concentration values for species NO for each
grid ceil along each of the boundary edges. The column numbers across the top indi-
cate either the x or y value, depending on the edge. The row numbers on the left
indicate the level number.
For presenting concentrations along the edges, the BOUNDARY program always
designates the west boundary = 1, east = 2, south = 3, and north = 4. These
designations are different from the ones used for user-defined boundary segments.
90008 20
336
-------
ooooo
y«.« — OO
H- O — O. —
-j X i_ ae
coo —— oo
O — —
—IOC fl
O< *-
QC O C
t— Z T3
o c uj
UJ UJ UJ QC
337
-------
— O. ^-
_l ffl —
• O O <"->
O OO «T
O •
00
1
n
I
I
EH
H
H
— — o
330
-------
•2OO OO
oo o o
ooo o
OOO OO
oooo
oooo
— «— -3
— — I O O — —
aaraooaa
Qj^aEot-j i '
o o i i x >•
OOOOO
oooo
ooo o
X >.
H
3
w
339
-------
LjJ Ul \ft U"» t/1
O1 U"> iTl Lft (^1
O ijj O
03
340
-------
— O T O O O O
o
o
O 03
o
o >—
0003 000-1000OOOOOOOOOOOOO
oooo ooo^oooooooooooooooo
OOOO OOO^OOOOOOOOOOOOOOOO
oooo ooo<-»oooooooooooooooo
0000 000~10030000000000030
OOOO OOO<*10OOOOOOOOO30000O
OOOO OOOflOOOOOOOOOOOOOOOO
-J10003 300-1-0000000000000000
o
ce^oooo ooonoooooooooooooooo
O 030-1000000000-3000000
33OOOOOO 333333
333^1
3 3 3 3
3333333 3 O3O3O 33
333330300000003
O 3 3 3 30
3 —
Ti a:
— J3
= <
3 Z
-5 3
-13333333333033303
O333 030-13033333333330030
oooo 0001-10000000000000000
— O ^r O O
O O O
o >-
03 ae
<
Z CO
<
o
_j a.
— <
— t^OOO OOOO3O3OOO3OOO333333
(NjOOOO coOOOOOOOOOOOOOOOOOOO
— oooo oooooooooooooooooooo
§
r-
CQ
M
341
-------
0303030
o 000000000000000
S,«oooooaoooo<*ioooooooooooooooo
•a < m
oooooooooooo
330
00333"000 000 0
000033000
^0000033
0000
3000300033030000
-,0033300300000003
0000^0000000000000000
00300003000~10000000
S~-OOOOOOOO-1000000O
000000000
33330300^
o oooooooooooo
ooooooooooooo
O -J
CO U4
(JO
342
-------
LjWUINLSS-\i i I
O
oorsiooooooooo
X
^,— 000000000
§00000-30000
CNJ
i.
O
^o^ooooooooo
O
O
o»ooooooooo
o^ooooooooo
— O O O O O
— O O O O O
— O O O O O
— O O O O
— O O O O O
— O O O O O
— 00000
— O O O O O
O O O O
-^ooooooooo
« —
O
0 '-1 OOOOOOOOO
_— 300000300
— O O O O O
— O O O O O
— O O O O O
— 00000
-00000
0000000000
•3 01
030033003 —
=' <
3 a:
~^ 3. voooooooooo — ooooo
m ae
^ujiriOOOOOOOOO^OOOOO
— z
"3
~^rrrrrr«r,T
-------
o
o
o
o
a
o
rooooooo ooooooooo
— OOOOO
GBaCOOOOOOOOOOO — 00000
— 03030
— 3OOOO
OOOOO3
— 3OOOO
000000030 —
o o o
—-OOOO
o o o
— 0000000000 — 00000
•OujOlOOOOOOOOO^-OOOOO
f— ooooo
000000000 — OOOOO
OCuj^OOOOOOOOOO — OOOOO
O — OOOOO
3 ~J
o -»
344
-------
OC ,—
h-0
ZO
c
I
CQ
345
-------
oooooooooooooo
oooooooooooooo'oooooooooooo
0000
oooooooo
oooooooooooooooooooo
<«<«•*««« ««««"*«^-t't«l'l=l^-"ir:ici-;-;^:i:ii:i-iiiii»»^>=»>»»»=>=>;>^:>:>::>:>:'
c.^-x.aa-xa-x.^ — i->^^x.x = <=*<-"="
346
-------
I I
iggggggggggggggggggggggggggggggg
OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
ggggggggggggggggggggggggggi§°§il
— oooooooao
oo
oo
oo
00
= y- I X G. O^*""!
j
-------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 o o oooooooooooooooooooooooooooooooooooooooo oooooo oooo
OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
OOOOOOOOOO'OOOOQOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOQO OOO O
OOOOOOOOOOOOOOOOOOOOOQOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOQOOOOO
OOOOOOOOOOOOOOOOOOOOOOOOOOOC3OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
I
O
CJ
xxxxxzxxxxxxxx
EH
H
m
H
$
w
348
-------
ooo 0000:20
•30003000000000000000000000000000
•3000000000000000000000 = 000000000
OOOO
2OOOOOOOOOQOOOOOOOOOOOOOOQOO
,
zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz
OOOOOO
OOOOOOOOOOOOOOO^OOOOOOOOOOOOOOOO
'OOrM'— co^noooocz OOOOCD^O ^
— OOO'— OOOOOOOOO' — O ClOd
ooooooooooocooooooooo^ooo
c
•f-l
a
CJ
a ZOO
—JOO <*!*— —t
!»L*j_i-^«tOZ:OOO ZOO
jocLao.zx<— i— JOOOCMZ
coiOQ.za.xoj — uji^z
CQ
349
-------
o
o
o
o —
o o
o o
~- CO <--"
o — o
O O O
000
o
o
o
— o o
303
O -J O
0 — 00
oo-^o
0000
Of O -J
O ^ at
— < O
O Q- —
O •* <
c
n
I
r-
fl
^
W
350
-------
— 33 -n
o — o
o o 3
ooo
0303
0 = 33
0033
•H
x-t
O
I
r-
H
CQ
H
W
351
-------
o
x
r-
i
r»
352
-------
. .ill
o *** ^
0
o
o o
o o
o o
o
o
o o o
000
003
0-590
o o o o o
o ^ o
— o o
o o
o o
w -- - -
— r^ m m
—i -- O
3 o o
—I O O O
— 000
0* 01 01
oe i • i
0.
^) tNj 3 3 3
— 333
<* 3 O O
< 3 o a
~ 01 0. 01
o
h- 000
01
0 0
0 0
0 0
* 0>
i i
fj>
TJ
>-
O
0
CD
(
X
uu
_J O O O O O
— o o o o o
o o o o o
— — 3 3 3 3 3
X. 3 O 3 3 3
- esj o o o o o
< o o o o o
oe o o o o o
z o o o o o
UJ
ae
o
o o
r^
c
13
>-
a
o
,
X
z
UJ
±
ae
a_
<
^i
0
<
2
i-j
Z
o
(_)
o
z
3 O O O
o o o o
01 01 0i 01
CD 3 3 3 3
~* O 3 3 3
3333
3330
01 01 01 01
m O O O o
o o o o
0000
01 01 01 01
-J
_J
a
— O 13
I
8
r-
0
W
353
-------
0.
*
•o
a.
a;
a.
••3
-r
-33
=
3
—5
r*1
C
TS
'
ae
o
z
o
CD
X
z
0
X
r
a.
o
0
3
O
*.
O
z
0
±
oc
a.
VI
o
QC
Y CONCENT
o
o
o
o
o
— 0
o
3
3
0
O
0
o
o
o
-g
o
0
o
3
=
3
0
Ol
T 3
0
O
o
Oi
3
3
0
— O
o
0
o
-J
_J
o
ae
0
o
0
o
o
o
o
o
o
o
0
o
3
o
o
o
3
3
3
en
3
•3
O
O
o
*
3
O
0
0
o
0
1
o
o
o
o
0
o
o
o
o
a
o
0
o
o
o
3
3
O
o
°
o
3
•x
3
O
O
o
?
o
o
o
0
o
o
1
0
o
a
o
o
3
o
o
o
0
o
g
°
o
0
°
3
3
o
o
o
en
0
3
0
O
o
0
o
o
0
o
o
o
o
3
o
o
0
o
o
°
g
o
1
3
-
o
01
§
0
0
o
o
'
a.
'It
-3
a.
-
ce
a.
3
3
"3
T
^
*
NDARY
0
CO
UJ
X
z
LJ.
— 1
>- UJ
O O
3 oc
O *J
CO
^ 3 O 3
0 O O 3
-------
fl <-> <•"
o o
o o o o o
o o o o o
O :3 O O O LU
',0 03 CO
— £ —'
0000
i_ ~. O
— f- O
0
0
o
t- ^- O
c
—I
r-
r-
PQ
M
g
w
355
-------
3 O 3
333
303
^
OJ
c
a
— 5
<*i
c
«i
-4-1
<
>-
it
<
O
(
UJ
X
z
JC
3
Z
O
~™
^
QC
Q-
ac
-
0
QC
LJ
Z
O
LJ
O
-
o
o
o
o
01
a
o
o
0
o>
3 a
O 3
o o
0^ • o
Z (X
0 Z
(-1
< 0
a
3
- m 3 O 3
^ 300
a. ...
z
o
< 000
oe ooo
z ooo
0
>- ^
o o
Ti Ti
3 0
o o
3 O
3 3
3 O
0 0
0 0
0 0
•8
a
H
m
H
-------
7.4 ALOFT BOUNDARY CONDITIONS FILE {TOPCONC)
The TOPCONC file contains time-varying matrices of poilutant concentrations at
the top of the region. Downward (negative) vertical wind speeds at the top of the
region will cause air parcels containing these poilutant concentrations to be trans-
ported into the region.
7.4.1 TOPCONC Preprocessor (TPCONC)
The TOPCONC file is created by the TPCONC preprocessing program (Figure 7-6).
TPCONC requires subroutines from the libraries UTILITY and FILUTIL. The pro-
gram reads a list of file names from unit 5 ("standard input") and writes some
diagnostic measures to unit 6 ("standard output"). Use of these two units is confined
to the program's main routine and the subroutine OPENTP.
7.4.2 TPCONC Input Format
The list of files read from unit 5 by TPCONC contains just five tines, each with a
single file and some comment text (Table 7-7). Ail five iines must be present; if a
file is not used, a dummy file name must be supplied.
The input data for the TPCONC program on unit 3 must be in the standardized for-
mat described in Chapter 4. This file must begin with a CONTROL packet followed
by a REGION packet. UNITS and STATIONS packets may follow. Within the first
TIME INTERVAL packet, the user must specify a method and vertical method for
each variable and supply the necessary data for the program to construct the file
using those methods.
The following methods for interpolating values of variables above ground, specified
in the METHOD packet (Section 4.2.9), can be used to generate the TOPCONC file:
90008 20
357
-------
(OUTBREAK
(REGIONTOP
'(12)
CList of
input and
>utput files
Input Data File
CONTROL
•
•
END
REGION
•
•
END
(+ other packets)
TOPCONC
(binary) \
FIGURE 7-6. Information flow for creation of the TOPCONC file.
EEE90008
358
-------
TABLE 7-7. Format of the TPCONC control file.
Line
Number
Columns
1-20
Columns
21-100
FORTRAN
Format
1
2
3
4
5
Blank or comment
Blank or comment
Blank or comment
Blank or comment
Blank or comment
Name of input file 20X, A80
Name of formatted output file 20X, A80
Name of DIFFBREAK file 20X, A80
Name of REGIONTOP file 20X, A30
Mame of output TOPCONC file 20X, A.80
90008 22
359
-------
CONSTANT
GRID VALUE
STATINTERP
POISSON
SPLIT/COMB
ABSTOPCONC
RELTOPCONC
E-WINTERP
N-SINTERP
USER
These methods are discussed in detail in Section 4.1.7.
The time span of the TOPCONC file must include the entire time span of the simula-
tion runs for which it is to be used. Concentrations at the top of the region are con-
sidered to be constant during each time interval.
The input packet structure for the TPCONC preprocessing program is shown in
Figure 7-7. Each of these packets is described in detail in Section 4.2. Following are
special incut packet considerations for the TOPCONC file.
CONTROL
The file name on line 2 must be TOPCONC'.
The control variables to be specified on lines 4 to 3 for TPCONC are shown in
Table 7-8.
The number of species must be greater than zero.
If there are input variables that do not appear as output variables, their number
must appear as Che number of user-defined variables. All such variables must
also be named in the UNITS packet.
If data from measuring stations are to be used (methods STATINTERP or
POISSON), the maximum number of such stations must be given.
The number of subregions must be at least one.
The maximum number of parameters must be sufficient to include all specifi-
cation of ail parameters.
90008 20
360
-------
CONTROL
END
REGION
END
f UNITS
Optional < I
LEND
^STATIONS
Optional <^ •
LEND
'TIME INTERVAL
r SUBREGION
i ' *
i'
I;END
Must appear j; METHOD
in the first
time interval
>END
' CONSTANTS
VEND
GRID VALUES
Include those packets appro-
priate for the method(s) selected...
At least one must appear
,END
• STATION READINGS
•END
; VERTICAL PROFILES
,END
VjENDTIME
Can be repeated
'r.nut lila MTUCIUI
361
-------
TABLE 7-8. Entries for the CONTROL packet for the
TOPCONC file.
Line
Number
Entry
Number of species
Number of user-defined variables
Number of stations
Number of subregions
Number of parameters
Spare
Output file number
Print input values
Print output grid
Spare
Spare
Spare
Print units table
Print station locations table
Print regional grid
Print methods cable
Print station values sable
Spare
Spare
Mumber of heights in profile
Spare
Print vertical profile tables
Spare
Spare
DIFFBREAK file number
REGIONTOP file number
Spare
Spare
Spare
Spare
90003 22
362
-------
The vertical controls (line 7) must contain the maximum number of profile
heights if ABSTOPCONC or RELTOPCONC is used. Otherwise, this line should
be blank.
The file unit assignment (line 3) must provide an entry for REGIONTOP if the
method ABSTOPCONC or RELTOPCONC is selected. It must also provide an
entry for DIFFBREAK if RELTOPCONC is selected.
The beginning and ending dates and times should reflect the time variation con-
siderations discussed above.
A set of output species names is required; their number must be the same as
the entry in the first control parameter on line 4.
REGION. This packet must follow the CONTROL packet. The vertical
parameters will be ignored for the TOPCONC file.
UNITS
This packet, if present, must follow the REGION packet. The UNITS packet
must be provided if:
Any input variable will be input in other than internal units.
Any user-defined variables are specified.
OORD or HEIGHT unit conversions are to be used.
The numder of user-defined variables must not exceed the maximum specified
.n the CONTROL packet.
STATIONS. This packet is required if the method STATINTERP or POISSON is
specified. The number of stations listed must not exceed the maximum speci-
fied in the CONTROL packet.
TIME INTERVAL. One or more TIME INTERVAL packets must be present. The
first time interval must begin at or before the beginning of thetime span speci-
fied on line 10 of the CONTROL packet. All time intervals must be contiguous
and of nonzero length. Each TIME INTERVAL packet contains one or more of
the following packets and ends with an ENDTIME card. Following the first
time interval, only those data that are to be changed need be specified.
SUBREGION. The first time interval must contain a SUBREGION packet; the
inclusion of this packet in other time intervals is optional. The number of sub-
regions must not exceed the maximum specified in the CONTROL packet.
90008 20
363
-------
METHOD. A method must be provided for every variable—including user-
defined variables—in every subregion in the first time interval. Methods can be
changed in subsequent TIME INTERVAL packets if desired. Note that each
parameter entry contributes to the overall parameter count; the total number
of parameters must not exceed the maximum specified in theCQNTROl packet.
CONSTANTS. If the method CONSTANT is assigned to any variable in the
METHOD packet, the first time interval must contain a CONSTANTS packet.
More than one CONSTANTS packet can appear in any time interval.
GRID VALUES. If the method GRID VALUE is assigned to any variable in the
METHOD packet, the first time interval must contain a GRID VALUES
packet. More than one GRID VALUES packet can appear in any time interval.
STATION READINGS, if either the POISSON or STATINTERP method is
assigned to any species in the METHOD packet, the first time interval must
contain a STATION READINGS packet. More than one STATION READINGS
packet can appear in any time interval.
VERTICAL PROFILES. If the method ABSTOPCONC or RELTOPCONC is
assigned to any species in the METHOD packet, the first time interval must
contain a VERTICAL PROFILES packet. There must be a vertical profile
defined (or implied by means oi ALL) for every variable in every suoregion for
which the profile method was specified. The number of height-value pairs in
any singie profile must not exceed :he maximum specified in the CONTROL
packet. More than one VERTICAL PROFILES packet can appear in any time
interval.
It the method ABSTOPCONC or RELTOPCONC is selected, the REGIONTOP file
will be read by TPCONC. In addition, if the method RELTOPCONC is selected, the
DIFFBREAK file will also be read.
The input file for the creation of the TOPCONC file for the example application is
shown in Exhibit 7-8. The example file sets the boundary conditions aloft for the
entire span of the June 3-4, 1984 simulation of the Atlanta region. The time span of
the file and the first and only time interval are both set to begin on January 1, 1980
and end on December 31, 1989 so that it will more than cover the span of the simula-
tion. The file begins with standard CONTROL and REGION packets. Since no UNITS
packet is included, the input values for all species will be expressed in internal units.
90008 20
364
-------
fr
TCSJ-T—•
— .no
>>O — O O O
00
I
EH
M
CQ
M
zzo^->oa
j_luj
-------
§
u
oo
.
zooo zoo
< uj I— i^OOCMZ
O. X t-j--(j X Z X
CQ
H
366
-------
IUHCUNG
A single subregion is defined for the entire region, and the method for all species is
set to CONSTANT. In the example ail cells are set to the same value for each
species, but they can also be set to different values using an interpolation method or
by defining subregions. The values for TOPCONC in the example represent rela-
tively clean values, and are the same as those for the fictitious stations in the
AIRQUALITY inputs and for the BOUNDARY inputs.
7.*.3 TPCONC Output
The output variables from TPCONC are the species named in the CONTROL
packet. Additional user-defined input variables (e.g., reactive hydrocarbons") can be
specified in the UNITS packet. The internal units for the concentrations of ail
species except aerosols (AERO) are parts per million (ppm); for AERO, the units are
micrograms per cubic meter (ug/m ). The standard names for reactive species
recognized by the UAM are listed in Table 5-2. If any of these species does not
appear on the TOPCONC file, the top concentrations used by the UAM will be set to
a value defined in the CHEMPARAM file. Any species that appear in TOPCONC
that are not defined in CHEMPARAM will be ignored.
The formatted output from the example this input file is shown in Exhibit 7-9.
0 0 0 3 ^ 0
367
-------
ooooo
— oooo
—00000
u_ v> ae
O
OJ
t— — a. z
>o-»ooo
_J uj O
_j -j m z
« —
«_» a: —i —i
LJ oe t— n
03 CD =1 Q-
x o a. t—
=j ae a: =
Z Q. — O
353
-------
TOPCONC
K- O UJ
— a. _j
-I < U-
UJ O U.
M UJ Q — «
00
O
O
- x x a. oc\j
-------
o
o
as z —
»— Q k- *- O w <
Z w Z O O 2
—• >— — as o ^
oc z ae oo o >i
CL —• ^. —J ("""i w<
as — (j* —
«—j Q. -j — oo «_>
CO CQ 2 UJ
»—