EPA-454/B-95-003c
ADDENDUM
USER'S GUIDE FOR THE
i
INDUSTRIAL SOURCE COMPLEX (ISC3) DISPERSION MODELS
VOLUME I - USER INSTRUCTIONS
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air Quality Planning and Standards
Emissions, Monitoring, and Analysis Division
Research Triangle Park, North Carolina 27711
April 2000
-------
ACKNOWLEDGMENTS
The Addendum to the User's Guide for the ISC3 Models has been prepared by
Roger W. Erode of Pacific Environmental Services, Inc., Research Triangle Park, North
Carolina, under subcontract to EC/R, Inc., Chapel Hill, North Carolina. This effort has
been funded by the Environmental Protection Agency under Contract No. 68D98006, with
Dennis G. Atkinson as Work Assignment Manager.
11
-------
USER INSTRUCTIONS FOR THE
REVISED ISCST3 MODEL (DATED 00101)
This document provides user instructions for recent enhancements of the ISCST3
model, including the most recent version dated 00101 (April 10, 2000). The enhancements
described in this Addendum include changes to the processing of multi-year averages for
post-1997 PM10 NAAQS analyses, enhancements to the model which were formerly
available in draft form as ISCST390 (dated 97365), enhancements to the model for air
toxics applications, and an option to specify variable emission rate factors that vary by
season, hour-of-day, and day-of-week. The enhancements from the draft ISCST390 model
include a conversion to Fortran 90 in order to make use of allocatable arrays for data
storage, incorporation of the EVENT processing from the ISCEV3 model, an INCLUDED
keyword option for the source, receptor and event pathways, and two new options for
specifying area sources. The use of allocatable arrays provides much more flexibility for
the end user of the ISCST3 model. The enhancements for air toxics applications include the
Sampled Chronological Input Model (SCIM) option, optimizations for the area source and
dry depletion algorithms, inclusion of the gas dry deposition algorithms based on the draft
GDISCDFT model (dated 96248), and the option to output results by season and hour-of-
day (SEASONHR). User instructions for these enhancements are provided below.
ENHANCEMENTS INTRODUCED WITH ISCST3 (DATED 98348)
Post-1997 PM10 Processing
A new NAAQS for modeling PM10 was promulgated in July 1997. This guidance
utilizes the expected second high value of the 24-hour NAAQS replaced by a 3-year average
of the 99th percentile value of the frequency distribution and a 3-year average of the annual
mean. Since the Guideline on Air Quality Modeling precludes the use of a 3-year data set,
a policy was established that uses unbiased estimates of the 3-year averages, utilizing all
meteorological data (both single and multiple years of data) available. An unbiased estimate
of the 99th percentile is the fourth highest concentration, if one year of meteorological data
are input to the model, or the multi-year average of the fourth highest concentrations, if
more than one year of meteorological data are input to the model. Similarly, an unbiased
estimate of the 3-year average annual mean is simply the annual mean, if only one year of
meteorological data are input to the model, or the multi-year average annual mean if
multiple years of meteorological data are used. Analogously to the original NAAQS
situation, the entire area is in compliance when the highest fourth high (or highest average
fourth high) and the highest annual mean (or the highest average annual mean) are less than
or equal to the NAAQS.
The revised ISCST3 model will process the 24-hour and annual averages for PM10
according to the new NAAQS if the pollutant ID specified on the CO POLLUTID card is
PM10 or PM-10, and the CO MULTYEAR card is not present. In this case, the model will
1
-------
compute an average of the fourth highest concentrations at each receptor across the number
of years of meteorological data being processed. For a single year of data, the model will
report the fourth highest concentration at each receptor. For a five year period of data, the
model will report the average of the five fourth-highest values at each receptor. Also, for
multiple year data files, the annual average will first be calculated for each individual year
of data, and the average of these across the number of years will be calculated. This
processing of the annual average across multiple years may give slightly different results
than the PERIOD average across the same time period, due to differences in the number of
calms from year to year. In order to accommodate this difference, the new PM10 NAAQS
makes use of the ANNUAL average keyword for specifying the long-term average.
i
Users should be aware of the following restrictions which are applied to the new
PM10 NAAQS processing.
1. The averaging periods are limited to the 24-hour and ANNUAL averages. Use of
the PERIOD average or a short-term average other than 24-hour will result in a fatal
error message being generated.
2. Only the FOURTH (or 4TH) highest value may be requested on the RECTABLE
card for 24-hour averages. Specifying another high value on the RECTABLE card
will result in a fatal error message being generated.
3. The model will only process complete years of meteorological data, although there is
no restriction on the start date for the data. If less than one complete year of data is
processed, a fatal error message will be generated. If additional meteorological data
remains after the end of the last complete year of data, the remaining data will be
ignored, and a non-fatal warning message will be generated specifying the number of
hours ignored.
4. The MULTYEAR card cannot be used with the new PM10 NAAQS. Multiple year
analyses should be accomplished by including the multiple years of meteorology in a
single data file.
5. Since the 24-hour average design values for post-1997 PM10 analyses may consist of
averages over a multi-year period, they are incompatible with the EVENT processor.
If the MAXIFILE option is used to output 24-hour average threshold violations,
these may be used with the EVENT processor. Therefore, if the EVENTFIL option
is used without the MAXIFILE option for post-1997 PM,0 analyses, a non-fatal
warning message will be generated, and the EVENTFIL option will be ignored.
The revised ISCST3 model may still be used to perform PM10 analyses according to
the pre-1997 NAAQS. This may be accomplished as before by use of the MULTYEAR
card on the CO pathway, except that the syntax for this keyword has been changed slightly.
The syntax and type are now as follows:
-------
Syntax:
CO MULTYEAR H6H Savfil (Inifil)
Type:
Optional, Non-repeatable
where H6H is a new secondary keyword that identifies this as a pre-1997 analysis, the Savfil
parameter specifies the filename for saving the results arrays at the end of each year of
processing, and the Inifil parameter specifies the filename to use for initializing the results
arrays at the beginning of the current year. The Inifil parameter is optional, and should be
left blank for the first year in the multi-year series of runs. Other than the additional
secondary keyword of H6H, the MULTYEAR card works the same as in previous versions
of ISCST3. A non-fatal warning message will be generated if the MULTYEAR card is used
for pre-1997 NAAQS analyses.
Memory Allocation
The revised ISCST3 model will allocate data storage as needed based on the number
of sources, receptors, source groups, and other input requirements, up to the maximum
amount of memory available on the computer being used. The minimum system
requirements for this version of the model are a 386 or higher processor with a math
coprocessor and at least 2 MB of extended memory.
The revised ISCST3 model uses allocatable arrays to allocate data storage at model
runtime rather than at compile time, as done by the previous version of ISCST3. The
ISCST3 model preprocesses the model runstream input file to determine the data storage
requirements for a particular model run, and then allocates the input data arrays before
processing the setup data. Once the setup processing is completed, the model allocates
storage for the result arrays. When allocating data storage, the ISCST3 model traps for
errors, e.g., not enough memory available to allocate. If the allocation is unsuccessful, then
an error message is generated by the model and further processing is prevented. If the
CO RUNORNOT NOT option is selected, the model will still go through all array
allocations so that the user can determine if sufficient memory is available to complete the
run. Also, an estimate of the total amount of memory needed for a particular run is printed
out as part of the first page of printed output.
The parameters that are established at model runtime are as follows:
NSRC = Number of Sources
NREC = Number of Receptors
NGRP '= Number of Source Groups
NAVE = Number of Short Term Averaging Periods
NVAL = Number of High Values by Receptor (RECTABLE Keyword)
NTYP = Number of Output Types (CONG, DEPOS, DDEP and WDEP)
NMAX = Number of Overall Maximum Values (MAXTABLE Keyword)
NQF = Number of Variable Emission Rate Factors Per Source
NPDMAX = Number of Particle Diameter Categories Per Source
-------
IXM = Number of X-coord (Distance) Values Per Receptor Network
IYM = Number of Y-coord (Direction) Values Per Receptor Network
NNET = Number of Cartesian and/or Polar Receptor Networks
NEVE = Number of Events for EVENT processing
In the case of NPDMAX, if no particle information is present in the input runstream, then
NPDMAX is set to 1, otherwise it is set to 20. Other parameters are set to the actual
numbers required for a particular model run.
I
A change has also been made that affects the length of filenames that may be
specified in the ISCST3 model input file. A new PARAMETER called ILEN_FLD has
been added to MODULE MAIN1 in MODULES.FOR, which is initially assigned a value of
80. This PARAMETER is now used to specify the maximum length of individual fields on
the input runstream image, and also to declare the length of all filename and format
variables. This includes the input and output filenames specified on the command line.
EVENT Processing ,.
The revised ISCST3 model incorporates the EVENT processing from the ISCEV3
model. Currently, ISCST3 can be run in either the original ISCST3 mode or in the ISCEV3
mode for a particular model run. The input requirements of each mode are the same as for
the original ISCST3 and ISCEV3 models, respectively. In other words, ISCST3 will accept
input files that have been setup for either ISCST3 or ISCEV3.
INCLUDED Option
The INCLUDED keyword option allows for the user to incorporate source, receptor,
and/or event data from a separate file into an ISCST3 model runstream file. Multiple
INCLUDED cards may be placed anywhere within the source, receptor and/or event
pathway, after the STARTING card and before the FINISHED card (i.e., the STARTING
and FINISHED cards cannot be included in the external file). The data in the included file
will be processed as though it were part of the runstream file. The syntax and type of the
INCLUDED keyword are summarized below:
Syntax:
Type:
SO INCLUDED Incfil
RE INCLUDED Incfil
EV INCLUDED Incfil
Optional, Repeatable
where the Incfil parameter is a character field of up to 80 characters (controlled by the
ILEN_FLD PARAMETER in MAIN1) that identifies the filename for the included file.
The contents of the included file must be valid runstream images for the applicable pathway.
If an error is generated during processing of the included file, the error message will report
-------
the line number of the included file. If more than one INCLUDED file is specified for a
particular pathway, the user will first need to determine which file the error occurred in.
AREAPOLY and AREACIRC Source Type Options
The ISCST3 model includes two new options for specifying area sources. These are
identified by the AREAPOLY and AREACIRC source types on the SO LOCATION
keyword. The syntax, type and order of the LOCATION keyword are summarized below:
Syntax:
SO LOCATION Srcid Srctyp Xs Ys (Zs)
Type:
Mandatory, Repeatable
Order* Must be first card for each source input
where the Srcid parameter is the alphanumeric source ID defined by the user (up to eight
characters), Srctyp is the source type, which is identified by one of the secondary keywords
- POINT. VOLUME. AREA. AREAPOLY. or AREACIRC - and Xs, Ys, and Zs are the x,
y, and z coordinates of the source location in meters. All three of the area source types use
the same numerical integration algorithm for estimating impacts from area sources, and are
merely different options for specifying the shape of the area source. The AREA source
keyword may be used to specify a rectangular-shaped area source with arbitrary orientation;
the AREAPOLY source keyword may be used to specify an area source as an irregularly-
shaped polygon of up to 20 sides; and the AREACIRC source keyword may be used to
specify a circular-shaped area source (modeled as an equal-area polygon of up to 20 sides).
Note that the source elevation, Zs, is an optional parameter. The x (east-west) and y
(north-south) coordinates are for the center of the source for POINT. VOLUME, and
AREACIRC sources, and are for one of the vertices of the source for AREA and
AREAPOLY sources. The source coordinates may be input as Universal Transverse
Mercator (UTM) coordinates, or may be referenced to a user-defined origin.
The main source parameters for the AREAPOLY and AREACIRC source types are
input on the SRCPARAM card, which is a mandatory keyword for each source being
modeled. These inputs are described below
i
AREAPOLY Source Type
The AREAPOLY source type may be used to specify an area source as an
arbitrarily-shaped polygon of between 3 and 20 sides (the number of sides allowed may be
increased by modifying the NVMAX and NVMAX2 parameters in MODULES.FOR). This
source type option provides the user with considerable flexibility for specifying the shape of
an area source. The syntax, type and order for the SRCPARAM card for AREAPOLY
sources are summarized below:
-------
Svntax* S0 SRCPARAM Srcid Aremis Relhgt Nverts (Szinit)
Type:
Mandatory, Repeatable
Order*
t^le LOCATION card for each source input
where the Srcid parameter is the same source ID that was entered on the LOCATION card
for a particular source, and the other parameters are as follows:
Aremis - area emission rate in g/(s-m2),
Relhgt - release height above ground in meters,
Nverts - number of vertices (or sides) of the area source polygon,
Szinit - initial vertical dimension of the area source plume in meters (optional).
As with AREA sources, the emission rate for the source is an emission rate per unit area,
which is different from the point and volume source emission rates, which are total emission
rates (g/s) for the source. The number of vertices (or sides) used to define the area source
polygon may vary between 3 and 20. The locations of the vertices are specified by use of
the AREA VERT keyword, which applies only to AREAPOLY sources. The syntax, type
and order for the AREA VERT keyword used for AREAPOLY sources are summarized
below:
Syntax:
SO'AREAVERT Srcid Xv(1)Yv(1) Xv(2) Yv(2)
Xv(I) Yv(I)
Type:
Mandatory for AREAPOLY sources, Repeatable
Order* Must
t'ne LncATI°N ar>d SRCPARAM card for each source input
where the Xv(I) and Yv(I) are the x-coordinate and y-coordinate values of the vertices of the
area source polygon. There must by Nverts pairs of coordinates for the area source, where
Nverts is the number of vertices specified for that source on the SRCPARAM card. The
first vertex, Xv(l) and Yv(l), must also match the coordinates given for the source location
on the LOCATION card, Xs and Ys. The remaining vertices may be defined in either a
clockwise or counter-clockwise order from the point used for defining the source location.
AREACIRC Source Type
The AREACIRC source type may be used to specify an area source as a circular
shape. The model will automatically generate a regular polygon of up to 20 sides to
approximate the circular area source. The polygon will have the same area as that specified
for the circle. The syntax, type and order for the SRCPARAM card for AREACIRC
sources are summarized below:
-------
Svntax* S0 SRCPARAM Srcld Aremis Relhgt Radius (Nverts) (Szinit)
Type:
Mandatory, Repeatable
Order* Must ^°'-'-ow t'le LOCATION card for each source input
where the Srcid parameter is the same source ID that was entered on the LOCATION card
for a particular source, and the other parameters are as follows:
Aremis - area emission rate in g/(s-m2),
'K
Relhgt - release height above ground in meters,
(
Radius - radius of the circular area in meters,
.\ t
Nverts - number of vertices (or sides) of the area source polygon (optional, 20
sides will be used if omitted),
Szinit - initial vertical dimension of the area source plume in meters (optional).
As with AREA sources, the emission rate for the source is an emission rate per unit area,
which is different from the point and volume source emission rates, which are total emission
rates (g/s) for the source.
ENHANCEMENTS INTRODUCED WITH ISCST3 (DATED 99155)
TOXICS Option
The revised ISCST3 model includes enhancements for air toxics applications. These
enhancements include the Sampled Chronological Input Model (SCIM) option, optimizations
for the area source and dry depletion algorithms, inclusion of the gas dry deposition
algorithms based on the draft GDISCDFT model (dated 96248), and the option to output
results by season and hour-of-day (SEASONHR). In order to utilize these enhancements,
the user must include the TOXICS keyword on the CO MODELOPT card. Since the
TOXICS option is a non-regulatory default option, the DFAULT keyword should not be
included on the MODELOPT card. If the DFAULT keyword is present on the
MODELOPT card, the DFAULT option will override the TOXICS option if it is present,
and any other enhancements dependent on the TOXICS option. The enhancements
associated with the TOXICS option are described below.
Sampled Chronological Input Model (SCIM) Option
t
If the non-default TOXICS option is specified, the user may also use the SCIM
option to reduce model runtime. The SCIM option can only be used with the ANNUAL
average option, and is primarily applicable to multi-year model simulations. The approach
used by the SCIM option is to sample the meteorological data at a user-specified regular
-------
interval to approximate the long-term (i.e., ANNUAL) average impacts. Since wet
deposition does not occur at regular intervals, the user can also specify a separate wet
sampling interval to reduce the uncertainty introduced by sampling for wet deposition. The
DEPOS option is ignored when SCIM is selected because, depending upon whether or not
the user selected the separate wet hour sampling, the dry deposition and wet deposition rates
can be based on different sets of sampled hours. Therefore, the annualized deposition rates
for the two types of deposition are calculated separately. For this reason, the user is advised
to calculate dry and wet deposition rates separately (using DDEP and WDEP, respectively)
and add the two to obtain the total deposition rate when the SCIM option is used. Studies
have shown that the uncertainty in modeled results introduced by use of the SCIM option is
generally lower for area sources than for point sources.
When only the regular sampling is selected, all hourly impacts (concentration, dry
deposition flux and the wet deposition flux) are calculated in the normal fashion for each
sampled hour. The annual average concentration is then simply calculated by dividing the
cumulative concentration for the sampled hours by the number of hours sampled (arithmetic
average), and the annual dry and the wet deposition fluxes are calculated by scaling the
respective cumulative fluxes for the sampled hours by the ratio of the total hours to the
sampled hours. The following illustrates the calculation of the ANNUAL impacts when
only the regular sampling is selected:
K,
c = £CS/NS
D = EDs(Nt/NJ
W = EWs (Nt/Ns)
where:
C,D,W = Calculated cone, dry flux and wet flux, respectively
Z)Cs,Z)D , £w = Cumulative impacts for the sampled hours
NS = Number of sampled hours
Nt = Total number of hours in the data period
When the wet hour sampling is also selected along with regular sampling, the
impacts are calculated slightly differently. The concentrations and the dry deposition fluxes
are based on the weighted contributions from the regular samples, modeled as dry hours,
and the wet hour samples. The regular samples consist of all the hours based on regular
sampling interval, but the effects of precipitation are ignored so that their contribution
represents only dry conditions, while the contribution from the wet hour samples represents
only wet conditions. The wet deposition fluxes are only based on the wet hour samples.
The following illustrates the calculation of the ANNUAL impacts when both the regular
sampling as well as the wet hour sampling are selected:
-------
Nt
D = EDd(Ntd/Nsd) .+
W = EWw(Ntw/NsJ
where:
C,D,W = Calculated cone, dry flux and wet flux, respectively
Ecd, EDd = Cumulative impacts for regular (dry) sampled hours
EC , ED , Ew = Cumulative impacts for sampled wet hours
Nsd = Number of regular sampled hours, modeled as dry
Nsw = Number of sampled wet hours
Ntd = Total number of dry hours in the data period
Ntw = Total number of wet hours in the data period
Nt = Total number of hours in the data period (Ntd+ Ntw)
To use the SCIM option, the user must include the SCIM and TOXICS keywords on
the CO MODELOPT card, and also specify the SCIM sampling parameters on the ME
SCIMBYHR card. The SCIM parameters on the SCIMBYHR card specify the starting hour
and sampling interval for the regular or dry sample, and also for the wet sample if used.
The syntax and type of the SCIMBYHR keyword are summarized below:
SvntaX" ME'SCIMBYHR NRegStart NReglnt NWetStart NWetlnt (Filnam)
Type:
Optional, Non-repeatable
where the NRegStart and NReglnt parameters specify the first hour to be sampled and the
sampling interval when performing the regular sampling, respectively, and NWetStart and
NWetlnt parameters specify the first wet hour to sample and the wet hour sampling interval,
respectively. Optionally, the user can create an output file by specifying the Filnam
parameter containing the meteorological data for the sampled hours (in the same format used
in the summary of the first 24 hours of data included in the main output file).
Although the ME SCIMBYHR is an optional card, it is required when using the
SCIM option. NRegStart is required to have a value from 1 through 24, i.e., the first
sampled hour must be on the first day in the meteorological data file. There are no
restrictions for NReglnt; however, NReglnt would generally be greater than 1. For
example, NReglnt could be based on the formula (24n+l), where "n" is the number of days
to skip between samples, in order to ensure a regular diurnal cycle to the sampled hours
(e.g., 25 or 49). NWetStart must be no greater than NWetlnt. An input of 0 (zero) for
NWetlnt indicates that the user has not selected the wet hour sampling.
-------
Optimized Area Source and Dry Depletion Algorithms
When the TOXICS option is specified, the area source and dry depletion integration
routines are optimized to reduce model runtime. This is accomplished by incorporation of a
2-point Gaussian Quadrature routine for numerical integration for some situations instead of
the Romberg numerical integration utilized in the regulatory default mode. In addition, for
area sources with dry depletion, another optimization option is available to reduce model
runtime by specifying the AREADPLT keyword on the CO MODELOPT card. When the
AREADPLT option is specified the model will apply a single "effective" depletion factor to
the undepleted area source integral, rather than applying the numerical integration for
depletion within the area source integral. If AREADPLT is selected, the DRYDPLT option
for non-area sources is automatically selected.
Gas Dry Deposition Algorithm
The revised ISCST3 model has the option to model the effects of dry deposition for
gaseous pollutants. In order to utilize this algorithm, the non-default TOXICS option must
be specified on the CO MODELOPT card. There are three new keywords on the CO
pathway and one new keyword on the SO pathway that are used for specifying inputs for the
gas dry deposition algorithm. The user has the option of specifying the deposition velocity
to be used with the CO GASDEPVD card, or allowing the model to calculate the deposition
velocities. If the user does not specify the deposition velocity with the GASDEPVD
keyword, then the state of vegetation must be specified with the CO VEGSTATE card, and
the source parameters for gas deposition must be specified with the SO GASDEPOS card.
The user also has the option to override certain default reference parameters through use of
the CO GASDEPRF card. The inputs for these keywords are described below. The use of
the gas dry deposition algorithm in ISCST3 also requires additional meteorological
parameters, which can be provided by the MPRM meteorological preprocessor. The
formats for the meteorological data input file for gas dry deposition applications is also
described below.
Specifying the State of Vegetation
An optional keyword is available on the Control pathway to allow the user to specify
the state of vegetation for use with the gaseous dry deposition algorithm of the ISCST3 model.
Three options are available on this keyword, one for active and unstressed vegetation, one for
active and stressed vegetation, and another for inactive vegetation.
The syntax and type of the VEGSTATE keyword are summarized below:
Syntax:
CO VEGSTATE UNSTRESSED or STRESSED or INACTIVE
Type:
Optional, Non-repeatable
10
-------
where the secondary keyword options describe the three options for the state of vegetation.
The state of vegetation is used in the model, along with ambient temperature and incoming
short-wave radiation, to determine the resistance to transport through the stomatal pores.
For unirrigated vegetation, the user should select the appropriate option for vegetation state
based on existing soil moisture conditions. For irrigated vegetation, the user should assume
that the vegetation is active and unstressed.
Option for Overriding Default Reference Parameters for Gas Dry Deposition
An optional keyword is available on the Control pathway to allow the user to
override the default reference parameters of cuticle resistance, ground resistance, and
pollutant reactivity foir use with the gas dry deposition algorithm.
The syntax and type of the GASDEPRF keyword are summarized below:-
Svntax* C0 GASDEPRF Rcutr Rgr Reactr (Refpoll)
Type:
Optional, Non-repeatable
where the parameter Rcutr is the reference value for cuticle resistance, Rgr is the reference
value for ground resistance, Reactr is the reference value for pollutant reactivity, and
Refpoll is the optional name of the reference pollutant. If the GASDEPRF keyword is
omitted, then the following default reference values for SO2 are used by the model: Rcutr
= 30 s/cm; Rgr = 10,-s/cm; and Reactr = 8.
Option for Specifying the Deposition Velocity for Gas Dry Deposition
An optional keyword is available on the Control pathway to allow the user to specify
the deposition velocity for use with the gaseous dry deposition algorithm of the ISCST3
model. A single deposition velocity can be input for a given model run, and is used for all
sources of gaseous pollutants. Selection of this option will by-pass the algorithm for
computing deposition velocities for gaseous pollutants, and should only be used when
sufficient data to run the algorithm are not available. Results of the ISCST3 model based on
a user-specified deposition velocity should be used with extra caution.
The syntax and type of the GASDEPVD keyword are summarized below:
Syntax:
CO GASDEPVD Uservd
Type:
Optional, Non-repeatable
where the parameter Uservd is the gaseous dry deposition velocity (m/s). A non-fatal
warning message is generated by the model if a value of Uservd greater than 0.05 m/s (5
cm/s) is input by the user. When the GASDEPVD keyword is used, the VEGSTATE and
11
-------
GASDEPRF keywords for the CO pathway, and the GASDEPOS keyword for the SO
pathway, are no longer applicable and cannot be used in the same model run.
.»
Specifying Source Parameters for Gas Dry Deposition
The input of source parameters for gas dry deposition is controlled by the
GASDEPOS keyword on the SO pathway. The gas dry deposition variables may be input
for a single source, or may be applied to a range of sources.
The syntax, type, and order for the GASDEPOS keyword are summarized below:
Svntax* S0 GASDEPOS Srcid (or Srcrng) Diff Alphas Reac Rsubm Henry
Type:
Optional, Repeatable
Order* Must ^°^ow t'le LOCATION carcl f°r each source input
where the Srcid or Srcrng identify the source or sources for which the inputs apply, the
parameter Diff is the molecular diffusivity for the pollutant being modeled (cm2/s), Alphas
is the solubility enhancement factor (a,) for the pollutant, Reac is the pollutant reactivity
parameter, Rsubm is the mesophyll resistance term (rj for the pollutant (s/cm), and Henry
is the Henry's Law coefficient for the parameter. Values of these physical parameters for
several common pollutants may be found in chemical engineering handbooks and various
publications, such as the Air/Superfund National Technical Guidance Study Series (EPA,
1993). The Alphas and Henry parameters are only used when applying the algorithm over a
water surface. If no water surfaces are present in a particular application, then dummy
(non-zero) values may be input for Alphas and Henry. The model converts the input units
for Diff to m2/s and Rsubm to s/m before being used in the computations.
Meteorological Formats for Gas Dry Deposition
Since the deposition algorithms require additional meteorological variables, the exact
format of ASCII meteorological data will depend on whether the dry and/or wet deposition
algorithms are being used. If the deposition algorithms are being used, then the unformatted
data file cannot be used. The order of the meteorological variables for the formatted ASCII
files and the default ASCII format are as follows when the CARD option is used:
12
-------
ASCII Meteorological Formats With the CARD Option
•j'
Variable
Year (last 2 digits)
Month
Day
Hour
Flow Vector (deg.)
Wind Speed (m/s)
Ambient Temperature (K)
Stability Class
(A=1,B=2, ...F=6)
Rural Mixing Height (m)
Urban Mixing Height (m)
Wind Profile Exponent
(CARD only)
Vertical Potential
Temperature Gradient :(K/m)
(CARD only)
Friction Velocity (m/s)
(Dry or Wet Deposition Only)
Monin-Obukhov Length (m)
(Dry or Wet Deposition Only)
Surface Roughness Length (m)
(Dry or Wet Deposition Only)
* ?•.
Incoming Short-wave Radiation (W/m2)
(Gas Dry Deposition Only)
Leaf Area Index
(Gas Dry Deposition Only)
Precipitation Code (00-45)
(Wet Deposition Only)
Precipitation Rate (mm/hr)
(Wet Deposition Only)
Fortran
, Format
12
12
12
12
F9.4
F9.4
F6.1
12
F7.1
F7.1
F8.4
F8.4
F9.4
F10.1
F8.4
F8.1
F8.3
14
F7.2
Columns
1-2
3-4
5-6
7-8
9-17
18-26
27-32
33-34
35-41
42-48
49-56
57-65
66-74
75-84
85-92
93-100
101-108
109-112
(93-96
without Gas
Dry Deposition)
113-119
(97-103
without Gas
Dry Deposition)
13
-------
The order and default format of the meteorological variables for the formatted ASCII files
without the CARD option are as follows:
ASCH Meteorological Formats Without the CARD Option
Variable
Year (last 2 digits)
Month
Day
Hour -. .
Flow Vector (deg.)
Wind Speed (m/s)
Ambient Temperature (K)
Stability Class
(A=1,B=2, ...F=6)
Rural Mixing Height (m)
Urban Mixing Height (m)
Friction Velocity (m/s)
(Dry or Wet Deposition Only)
Monin-Obukhov Length (m)
(Dry or Wet Deposition Only)
Surface Roughness Le'ngth (m)
(Dry or Wet Deposition Only)
Incoming Short-wave Radiation (W/m2)
(Gas Dry Deposition Only)
Leaf Area Index
(Gas Dry Deposition Only)
Precipitation Code (00-45)
(Wet Deposition Only)
Precipitation Rate (mm/hr)
(Wet Deposition Only)
Fortran
Format
12
12
12
12
F9.4
F9.4
F6.1
12
F7.1
F7.1
F9.4
F10.1
F8.4
F8.1
F8.3
14
F7.2
Columns
1-2
3-4
5-6
7-8
9-17
18-26
27-32
33-34
35-41
42-48
49-57
58-67
68-75
76-83
84-91
92-95
(76-79
without Gas
Dry Deposition)
96-102
(80-86
without Gas
Dry Deposition)
14
-------
Season by Hour-of-Dav Output Option (SEASONHR)
When the non-default TOXICS option is specified, the user may request an output
file containing the average results (CONG, DEPOS, DDEP and/or WDEP) by season and
hour-of-day. To select this option, the user must include the SEASONHR keyword on the
OU pathway. The syntax, type, and order for the SEASONHR keyword are summarized
below:
Syntax:
Type:
OU SEASONHR GroupID
Optional, Repeatable
FileName (FileUnit)
where the GroupID parameter specifies the source group to be output, FileName specifies
the name of the output file, and the optional FileUnit parameter specifies an optional file
unit and must be greater than 20. If FileUnit is left blank, then the model will dynamically
assign a file unit based on the formula 302+IGRP*10, where IGRP is the group index
number. A sample from a SEASONHR output file is shown below:
* ISCST3 (99155): Example of
* MODELING OPTIONS USED:
* CONC WDEP
* FILE OF SEASON/HOUR
SEASONHR Output File Option
RURAL FLAT TOXICS
VALUES FOR SOURCE GROUP: ALL
WETDPL
* FOR A TOTAL OF 216 RECEPTORS.
* FORMAT:
* X
*
8.68241
17.36482
86.82409
173.64818
868.24091
1736.48181
17.10101
34.20201
171.01007
342.02014
1710.10071
3420.20142
25.00000
50.00000
250.00000
500.00000
2500.00000
5000.00000
(4(1X,F13.5)
V
49.24039
98.48077
492.40387
984.80774
4924.03857
9848.07715
46.98463
93.96926
469.84631
939.69263
4698.46289
9396.92578
43.30127
86.60254
433.01270
866.02539
4330.12695
8660.25391
,1X,F8.2,2X,A8,
AVERAGE CONC
0.00000
0.00000
0.18098
2.52520
2.07470
0.93252
0.00000
0.00000
0.15772
2.48554
6.09119
4.49830
0.00000
0.00000
0.10114
2.12970
2.79993
1.97200
2X,I4,2X,I4,
UET DEPO
0.00603
0.00177.
0.00008
0.00001
0.00000
0.00000
0.00002
0.00000
0.00000
0.00000
0.00000
0.00000'
0.000 17'
0.00001
0.00000
0.00000
0.00000.
0.00000
2X,I4,2X,
ZELEV
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
A8)
GRP
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
NHRS SEAS
87 1
87
87
87
87
87
87 1
87 1
87 1
87 1
87 1
87 1
87 1
87 1
87 1
87 1
87 1
87 1
HOUR NET ID
1 POL1
POL1
POL1
POL1
POL1
POL1
POL1
POL1
POL1
1 POL1
1 POL1
1 POL1
1 POL1
1 POL1
1 POL1
1 POL1
1 POL1
1 POL1
The NHRS column in the output file contains the number of non-calm and non-missing
hours used to calculate the season-by-hour-of-day averages. The SEAS column is the season
index, and is 1 for winter, 2 for spring, 3 for summer and 4 for fall. The records loop
through hour-of-day first, and then through the seasons.
15
-------
ENHANCEMENTS INTRODUCED WITH ISCST3 (DATED 00101)
Removal of UNFORM Option for Meteorological Data
The unformatted (binary) meteorological data option (ME INPUTFIL UNFORM) is
no longer supported by the ISCST3 model. Unnecessary code has been removed, and
proper error handling has been implemented. Users with unformatted meteorological data
should first convert the data to an ASCII format using the BINTOASC utility program
available on the SCRAM website. The unformatted data file option has been removed due
to unformatted files are not portable across different computer systems and compilers, and
unformatted files cannot be used with the deposition algorithms in ISCST3.
Season by Hour-of-Day and Day-of-Week Emission Factors
The variable emission rate factor option controlled by the EMISFACT keyword on
the SO pathway has been modified to include an option to specify variable emission rate
factors that vary by season, hour-of-day, and day-of-week. The day-of-week variability
allows for different emission factors to be specified for Weekdays (Monday-Friday),
Saturdays, and Sundays.
The syntax, type and order of the EMISFACT keyword are summarized below:
SvntaX' SO EMISFACT Srcid (or Srcrng) Qflag Qfact(i),i=1,n
Optional, Repeatable
Order* Must follow the LOCATION card for each source input
where the Srcid parameter is the same source ID that was entered on the LOCATION card
for a particular source. The user also has the option of using the Srcrng parameter for
specifying a range of sources for which the emission rate factors apply, instead of
identifying a single source. This is accomplished by two source ID character strings
separated by a dash, e.g., STACK1-STACK10.
The parameter Qflag is the variable emission rate flag, and is one of the following
secondary keywords: '
SEASON emission rates vary seasonally (n=4),
MONTH emission rates vary monthly (n=12),
HROFDY emission rates vary by hour-of-day (n=24),
STAR emission rates vary by speed and stability category (n=36),
SEASHR emission rates vary by season and hour-of-day (n=96), and
16
-------
SHRDOW emission rates vary by season, hour-of-day, and day-of-week [M-F,
Sat., Sun.] (n=288)
The Qfact array is the array of factors, where the number of factors is shown above for each
Qflag option. The EMISFACT card may be repeated as many times as necessary to input
all of the factors, and repeat values may be used for the numerical inputs. An example of
each of these options is presented below, with column headers to indicate the order in which
values are to be input.:
** . WINTER SPRING SUMMER.. FALL
SO EMISFACT STACK1 SEASON 0.50 0.50 1.00 • 0.75
** JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
SO EMISFACT STACK1 MONTH 0.1 0.2 0.3 0.4 0.5 0.5-0.5 0.6 0.7 1.0 1.0 1.0
1
6'7 8 9 10 11 12
SO EMISFACT STACK,! HROFDY 0.0 0.0 0.0 0.0 0.0 0.5 1.0 1.0 1.0 1.0 1.0 1.0
** 13 14 15 16 17 18 19 20 21 22 23 24
SO EMISFACT STACK1 HROFDY 1.0 1.0 1.0 1.0 1.0 0.5 0.0 0.0 0.0 0.0 0.0 0.0
** or, equivalently:
SO EMISFACT STACK1 HROFDY
1-5 6
5*0.0 0.5
7-17 18 19-24
11*1.0 0.5 6*0.0
** Stab. Cat.:
SO EMISFACT STACK1 STAR
A B C D E F (6 WS Cat.)
6*0.5 6*0.6 6*0.7 6*0.8 6*0.9 6*1.0
SO EMISFACT STACK1 SEASHR
enter 24 hourly scalars for each of the four
seasons (winter, spring, summer, fall), e.g.,
**
so
EMISFACT
STACK!
SEASHR
Winter
24*0.50
Spring
24*0.50
Summer
24*1.00
Fall
24*0.
75
SO EMISFACT STACK1 SHRDOW
** Weekdays:
SO EMISFACT STACK1 SHRDOW
** Saturdays:
SO EMISFACT STACK1 SHRDOW
** Sundays:
SO EMISFACT STACK1 SHRDOW
enter 24 hourly scalars for each of the four
seasons (winter, spring, summer, fall), first
for Weekdays (Monday-Friday), then for Saturdays,
and finally for Sundays, e.g..
Winter
24.1.0
Spring
24*0.8
Summer
24*0.6
Fall
24*0.8
24*0.5 24*0.4 24*0.3 24*0.4
24*0.25 24*0.2 '24*0.15 24*0.2
References
Environmental Protection Agency, 1993: Air/Superfund National Technical Guidance
Study Series, Models for Estimating Air Emission Rates from Superfund Remedial
Actions. EPA-451/R-93-001, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina 27711.
17
------- |