Improvements in Indirect Exposure Assessment Modeling: A Model for Estimating Air Concentrations and Deposition


EPA/600/A-94/060
Improvements in Indirect Exposure Assessment Modeling:
A Model for Estimating Air Concentrations and Deposition
Donna B. Schwede1
Atmospheric Sciences Modeling Division
Air Resources Laboratory, NOAA
Research Triangle Park, NC 27711
Joseph S. Scire
Sigma Research Corporation
196 Baker Avenue
Concord, MA 01742
'On assignment to the Atmospheric Research and Exposure Assessment Laboratory, U. S.
Environmental Protection Agency.

-------
94-TA23.05
INTRODUCTION
In 1990 the U.S Environmental Protection Agency (U.S. EPA) released the document "Methodology
for Assessing Health Risks Associated with Indirect Exposure to Combustor Emissions"1. The
purpose of this document is to provide guidance on conducting indirect exposure assessments. The
air model recommended in this document is the COMPDEP model. The COMPDEP model v.bs
developed in recognition of the need for an air model for predicting concentrations and deposition
flux, specifically in areas of complex terrain. The model was created by modifying COMPLEX I, a
U.S. EPA screening model2 for assessing pollutant impacts in complex terrain, to include existing
algorithms for modeling dry and wet deposition and building effects. To account for dry deposition,
the plume depletion algorithms from the Multiple Point Source Algorithm with Terrain Adjustments
Including Deposition and Sedimentation (MPTER-DS) model3 and the dry deposition velocity
algorithms from the California Air Resources Board (CARB)4 were implemented The algorithms for
determining the wet deposition flux were based on methods described in Slinn5 and PEI and Cramer6
which is a scavenging coefficient approach. The building wake effects algorithm from the Industrial
Source Complex - Short Term (ISCST)7 was also added for use in simple terrain.
In Spring 1993, the Agency convened a multidisciplinary workgroup to rwiew and update current
guidance on pa-forming indirect exposure assessments and in particular the 1990 methodology
document. To address the specific technical issues, the workgroup was divided into three subgroups:
air modeling and emissions, food chain modeling, and human exposure and contact rates. After
considerable review, the air modeling and emissions subgroup recommended that the COMPDEP
model be updated
The areas of the COMPDEP model that were identified for improvement include dry deposition of
particles and gases, wet deposition of gases and particles, building wake effects, deposition in
complex terrain, and modeling of multiple point sources and area and volume sources. The ISC-
COMPDEP model was developed to address these concerns. ISC-COMPDEP was developed from
the latest version of the ISCST model, now referred to as ISCST28. This platform was selected over
COMPLEX I because ISCST2 already contained better handling of building wake effects and the
capability to model multiple point sources as well as area and volume sources. Also, a review of
particle diy deposition algorithms has just been completed9 and an improved particle dry deposition
algorithm was recently implemented into ISCST2. The area source algorithm was also recently
updated10. The first priority was to add the complex terrain and wet deposition modules to ISCST2.
Longer-term goals include addressing deposition of gases and scavenging coefficients used in
calculating the wet deposition flux. In the sections that follow, we discuss the major differences
between COMPDEP and ISC-COMPDEP.
DISCUSSION OF MODEL ALGORITHMS
Many aspects of COMPDEP and ISC-COMPDEP are the same since they are boft designed to
produce the same results as ISCSY for simple terrain and the same results as COMPLEX I for
complex terrain. They also both incorporate the U.S. EPA's intermediate terrain practice wherein
receptors between stack top and plume height are modeled using both the simple terrain and complex
terrain algorithms and the higher of the two estimates is retained for each hour. Some differences
between the models arise from their design. COMPDEP was built from COMPLEX I with some
2

-------
94-TA23.05
1SCST features added. Also, 1SCST has beer, upgraded to ISCST2 since the time of the original
development of COMPDEP in 1990. Modifications were made to model algorithms, in particular
those that handle the building wake effects as well as other model features. Therefore, not all of the
options currently available in ISCST2 are available in COMPDEP. For example, COMPDEF only
contains rural dispersion coefficients whereas ISCST2 contains both urban and rural coefficients.
Also, COMPDEP is only used for modeling elevated point sources, while 1SC-COMPDEP can be
used for point, area, or volume sources. Conversely, ISC-COMPDEP was created from the ISCST2
platform with COMPLEX I algoritlims added. COMPLEX I was implemented in the "regulatory
default mode" so that run-time options are autorratically set to agree with the recommendation in the
Guideline on Air Quality Models (Revised)2. Therefore, the complex terrain calculations in ISC-
COMPDEP will always use the rural dispersion coefficients, rural mixing heights, and gradual plume
rise with no regard to the settings of these options in the input file.
The differences between the models that can be expected to cause the greatest impact on
concentration and deposition ttux are in the algorithms that handle building effects, dry deposition,
and wet deposition. These areas are discussed in greater detail below.
Building Effects
In both COMPDEP and ISC-COMPDEP, users have the option of modeling the effects of buildings
cn plume dispersion for simple terrain calculations. The models calculate the distance dependent
momentum plume rise at a downwind distance of two building heights. The plume is considered to
be affected by the building if the plume height is less than 2.5 times the building height or less than
the sum of the building height and 1.5 times the building width. COMPDEP contains the Huber-
Snyder building effects algorithms11 which modifies the lateral and vertical dispersion parameters (Cy
and qj to account for the effects of buildings. ISC-COMPDEP contains both the Huber-Snyder and
Schulman-Scire12 wake effects procedures. The Schulman-Scire procedure is invoked when the
stack height is less than the sum of the building height and 0.5 times the minimum of the building
height and the building width. The Schulman-Scire procedure uses a modified plume height that
includes the effects of the initial dilution of the plume with ambient air and enhances c^as a function
of plume height. The differences between these approaches is a particularly important consideration
in modeling short stacks. A second difference between the wake effects algorithms in COMPDEP
and ISC-COMPDEP is th2l ISC-COMPDEP allows for direction specific building dimensions
whereas a single height and width are input to COMPDEP.
Dry Deposition
The calculation of the diy deposition and settling flux of a pollutant requires the specification of the
deposition velocity, settling velocity, and the near-surface concentration of the pollutant The
concentration estimate includes the effects of plume depletion due to deposition. COMPDEP and
ISC-COMPDEP use different approaches to obtain these values and therefore will provide different
estimates of the dry deposition flux. COMPDEP uses the CARB model4 for determining the
deposition velocity and the K-theoiy plume depletion method described in Rao13. ISC-COMPDEP
contains the dry deposition algoritlim recently implemented in ISCST2. The deposition velocity is
calcilatec' using a modified form of the method in the ADOM model9,14. Piune depletion is
accounted for by using the Horst modified source depletion method 15. These methods are discussed
in this sea ion.
3

-------
94-TA23.05
Many of the methods for determining the deposition velocity use a resistance approach where the
velocity is a function of the resistance to movement of the pollutant through the atmosphere. Both
COMPDEP and ISC-COiMPDEP use resistance models to calculate the deposition velocity.
COMPDEP uses the CARB model4 for calculating the deposition velocity of a particle which is
based on the work of Sehmel and Hodgson16. In this method, the atmosphere is divided into two
layers, each characterized by a different resistance. The atmospheric diUjsional resistance
characterizes the upper layer (one meter above the surface to one centimeter above the surface).
Atmospheric turbulence is the dominant transfer mechanism in this layer. In the lower layer, the
importance of Brownian motion is accounted for in the surface resistance integral. The relationships
for determining the deposition velocity were derived from an empirical fit to wind tunnel data. In
COMPDEP, the deposition velocity, vd, is calculated as follows:
V -
f-v,(VW	(I>
1. - exp
u_
where
vd	=	deposition velocity (cm/s),
v4	=	gravitational settling velocity (cm/s),
In	=	atmospheric diffusional resistance integral (dimensionless),
13	=	surface resistance integral (dimensionless), and
u.	=	surface friction velocity (cm/s).
The gravitational settling velocity is a function of the particle density and the square of the particle
diameter. As particle diameter increases, gravitational settling dominates the removal process and the
deposition velocity approaches the gravitational settling velocity. The atmospheric diflusicnal
resistance integral includes the effects of eddy diffusion and Brownian motion. This integral is
evaluated using the flux-profile relationships of Businger et al17. The integral is largest for stable
atmospheres and least for unstable atmospheres. The surface resistance integral is based on an
empirical least-squares fit to deposition velocities of monodispersed particles to four different
surfaces observed in wind-tunnel experiments. The integral is calculated as a function of particle
diameter, surface roughness, and Brownian diffusion coefficient.
The ISC-COMPDEP model also divides the atmosphere into two layers for the purpose of calculating
the deposition velocity. The turbulent upper layer is characterized by the aerodynamic resistance.
The quasi-laminar lower layer is characterized by the deposition layer resistance. The deposition
velocity contains explicit parameterizations of these resistances plus gravitational settling terms14,18:
v, = 			 + v.	(2)
' + rd + V*v, *
where
vd = deposition velocity (cm;s),
vg = gravitational settling velocity (cm/s),
4

-------
94-TA23.05
ra = aerodynamic resistance (s/cm), and
rd = deposition layer resistance (s/cm).
In ISC-COMPDEP, the gravitational settling velocity is calculated similarly to the method in
COMPDEP (i.e. Stokes Law), however the correction factor that accounts for non-continuum effects
for small particles is slightly different. The two equations yield essentially the same gravitational
settling velocities. The method for calculating the aerodynamic resistance in ISC-COMPDEP is
based on Wesely and Hicks19 and is a function of wind speed, stability, and surface roughness. The
aerodynamic resistance is comparable in function to the atmospheric difiusional resistance integral in
COMPDEP, however the forniulaiicns are not equivalent The parameterization of the deposition
layer resistance is based on Pleim et al N. This parameterization uses the Schmidt number to
account for the effects of Brownian motion and the Stokes number which indicates the importance of
inertial impaction.
Figure 1 shows an example comparison of the deposition velocities predicted by COMPDEP and
ISC-COMPDEP for a range of particle sizes. For all but the smallest particles, the deposition
velocities predicted by COMPDEP are higher than those predicted by ISC-COMPDEP. The greatest
difference occurs for particle diameters around 1-2 nm. The deposition velocity approaches the
gravitational settling velocity at larger particle diameters for both models.
Both of these approaches to calculating the deposition velocity require an estimation of the friction
velocity (u.) and the Monin-Obukhov length (L). In COMPDEP, the user specifies one value for the
surface roughness length. The Monin-Obukhov length is determined from a linear approximation to
Golder's stability curves20. The friction velocity is then determined from the wind speed, Zq, and L
using the integral form of Businger's flux profile relationships as described in McRae21. To calculate
L and u. for use with ISC-COMPDEP, a separate preprocessor (DEPMET) is provided The
preprocessor requires the user to input the following site specific parameters: surface roughness,
displacement height, noon-time albedo, soil moisture availability parameter, fraction of the net
radiation absorbed at ihe ground, anthropogenic heat flux, and minimum value for L. The values of
these parameters can be varied by hour or held constant for all hours. Tables are provided in the
documentation associated with ISCST2 to suggest values for these parameters. To determine u. and
L, the sensible heat flax is first derived using the methods of Holstiag and van Ulden22. The friction
velocity is determined from the logarithmic wind profile with extensions for neutral and unstable
conditions as in Wang and Chen23. Finally, L is calculated from its definitioa
Table 1 shows a comparison of the values of L and u. calculated by these two methods for different
stability classes. In general, the friction velocities predicted by DEPMET are slightly higher than
those estimated by COMPDEP. Figure 2 is a plot of deposition velocity as a function of friction
velocity for particle sizes of 1 and 10 urn Both models are more sensitive to u. at lower values.
Therefore, predicted deposition velocities for low wind speeds and smooth surfaces could be affected
by the method used to estimate u.. The Monin-Obukhov values shown in Table 1 are generally
within one stability class. Neither model is sensitive to L, so the differences in the values of L
predicted by DILPMET arid COMPDEP are not significant
The final aspect that must be accounted for in determining tne dry deposiiion flax is depletion of the
plume due to deposition. COMPDEP contains the Rao13 K-theory model for calculating
concentrations considering the effects of depletion due tr upwind deposition The Rao approach
5

-------
94-TA23.05
consists of analytical solutions of the gradient transfer equation that are extensions of the traditional
Gaussian plume diffusion equation. One of the assumptions made to achieve the solution is that the
eddy difftisivity (K) is constant with downwind distance (x). This solution is only strictly valid when
the standard deviation of the concentration in the y and z directions varies as &s. For some cases,
(primarily at large distances), the K-theory model may not conserve mass exactly.
ISC-COMPDEP uses the modified source depletion approach developed by Horst15 and extended for
stability classes A and B as in U.S. EPA9 to account for plume depletion. The key component of
this approach is the use of a profile correction factor to adjust the diffusion function. This approach
recognizes that, as material is deposited from a plume, most of the mass is lost from the lower
portion of the plume so that the vertical profile becomes non-Gaussian. The profile correction factor
depends on the history of the deposition flux and the rate of turbulent transfer.
Both the Rao and Horst methods were developed without consideration for elevated terrain.
However, as implemented in COMPDEP and ISC-COMPDEP, the presence of elevated terrain does
affect the copantration and deposition flux. In both models, the plume height is adjusted to account
for terrain u, • > , which generally results in a higher concentration and deposition flux. In ISC-
COMPDEP, the elle-.ts of elevated terrain arc also important in the calculation of the plume
depletion factor. The plume depletion calculation in ISC-COMPDEP involves the integration of the
deposition flux along the plume trajectory. Changes in terrain elevation between the source and the
receptor will affect the resulting factor. The user has the option of using gridded terrain data to
allow for a truer calculation of the integral or a simple linear slope approximation can be used if
gridded terrain data are not available.
Wet Deposition
The methods for calculating the wet deposition flax are slightly different in COMPDEP and ISC-
COMPDEP. Both models use a scavenging coefficient approach where the flux is calculated as the
product of the scavenging coefficient and the vertically integrated concentration:
.1 /
fw = AM*Le
where
Fw =
wet deposition flux (g/Ws)
A
scavenging coefficient (s!)
Q(x) =
emission rate ($'i)
u
wind speed (m's)
=
standard deviation of the lateral plume dispersion (m)
y
crosswind distance (m)
The main difference between the i.;odels is in the inclusion of an "F-value" in the numerator of
equation (3) in COMPDEP which represents the fraction of the hour during which precipitation
occurred. Using the F-value approach, wet deposition occurs for the fraction of the hour that
precipitation occurred and dry dqwsition occurs for the remainder of the hour. In ISC-COMPDEP,
6

-------
94-TA23.05
dry deposition occurs continuously with wet deposition occurring for the entire hour whenever
precipitation is reported during the hour. Another difference between the models is in the
specification of the scavenging coefficient In COMPDEP, scavenging coefficients are specified by
particle size category and by precipitation intensity category. Hie three standard National Weather
Service precipitation intensity categories are used: light (trace to 0.10 m/h), moderate (0.11 to 0.30
in/h), and heavy (greater than 0.30 in/h). Only one set of coefficients is input and is assumed to be
for liquid precipitation; frozen precipitation is not modeled. For ISC-COMPDEP, users input a
normalized scavenging ratio for each particle size which is defined for a base precipitation rate of 1
mm/h. This normalized ratio is multiplied by the actual precipitation rate for the hour to obtain the
scavenging coefficient. Separate normalized scavenging coefficients can be input for liquid and
frozen precipitation and for particles and gases.
Input Requirements
COMPDEP and ISC-COMPDEP require similar input information, however the format of the input
files are quite different. There is no user's guide for COMPDEP. A detailed description of the
format of the input files for COMPDEP is provided in the document "Overview to the COMPDEP
Model" which can be downioaded from the U.S. EPA's Support Center for Regulatory Air Models
(SCRAM) bulletin board system. Detailed information about the input files for ISC-COMPDEP can
be found in the user's manuals for ISCST2 and also in documentation provided on SCRAM.
Both models require a control file that includes information on runtime options, source parameters,
receptor locations, particle size distributions, and scavenging coefficients. The COMPDEP file is
similar in format to that used by COMPLEX I with a few additional switches and parameters for
modeling deposition and building effects. Only co-located point sources can be modeled by
COMPDEP. Up to 400 receptors may be input using the cartesian coordinate system, a maximum
of 10 particle sizes can be defined. The control file for ISC-COMPDEP uses the keyword format as
in ISCST2. Additional keywords have been added to the ISCST2 format to control the processing
for deposition and complex terrain. Multiple point sources, area and volume sources can be modeled
with ISC-COMPDEP, however only point sources can be modeled when the complex terrain
algorithms are invoked. Gridded and discrete receptors can be input in cartesian or polar
coordinates. A maximum of 20 particle sizes can be input Users have the option of using gridded
terrain data for use with the plume depletion algorithm. The terrain is input via a separate file.
Meteorological data is first preprocessed by MPRM or RAMMET to produce a binary file for use
with both models. The MPRM (or RAMMET) file must be produced in binary format for input
directly to COMPDEP. Precipitation data is input to COMPDEP via a separate file that contains the
F-value and precipitation category for each hour. A preprocessor reads the National Weather Service
CD-144 and TD-3240 files and creates the binary precipitation file required by COMPDEP. For
ISC-COMPDEP, two additional preprocessors must be run after MPRM (or RAMMET) to create the
meteorological file. A precipitation preprocessor (PMERGE25) provides a file of hourly precipitation
rate from the TD-3240 file. The precipitation file, MPRM (or RAMMET) file (binary or ASCII),
and the CD-144 file are then input to DEPMET which produces an ASCII file for input to ISC-
COMPDEP.
ISC-COMPDEP provides greater flexibility than COMPDEP in the types of output that can be
produced. The output from COMPDEP consists of a single file that contains a verification of the
input parameters and provide; a listing of the annual average concentration, total annual dry
7

-------
94-TA23.05
deposition flux, total annual wet deposition flux, and total annual combined deposition flux at each
receptor. With ISC-COMPDEP, the user may choose to output concentration or dry deposition flux
or wet deposition flux or total deposition flux. Multiple runs are required to obtain all of these
estimates. As with ISCST2, many types of output files can be produced including plot files and
avei aging time summaries.
CONCLUSIONS
ISC-COMPDEP represents an improvement in techniques for indirect exposure assessment modeling.
Technical improvements include modifying several key model algorithms including those that
account for building wake effects, dry deposition, and wet depositioa The modifications to the
building wake effects algorithms will be particularly important for modeling short stacks which arc
typical of the incinerators included in many risk assessments. The new dry deposition algorithm
which was found to perform well in a recent evaluation study9, was used in ISC-COMPDEP. The
wet deposition algorithm allows for an improved specification of scavenging coefficients. The input
requirements of ISC-COMPDEP are comparable to COMPDEP, however ISC-COMPDEP offers
greater flexibility in the types of output that can be generated.
Future versions of ISC-COMPDEP should include the capability to model the dry deposition of gns.-y
in addition to particles. Additionally, a more refined approach to modeling in complex terrain should
be investigated.
ACKNOWLEDGMENTS
The authors would like to acknowledge the contributions to the development of ISC-COMPDEP by
EPA Region V and the members of the air modeling and emissions subgroup of the Agency's
Indirect Exposure Workgroup.
DISCLAIMER
This paper has been reviewed in accordance with the U.S. Environmental Protection Agency's peer
and administrative review policies and approved for presentation and publication. Mention of trade
names or commercial products does nui constitute endorsement or recommendation for use.
REFERENCES
1. U.S. Environmental Protection Agency, Methodology for Assessing Health Uisks Associated wirh
Indirect Exposure to Combustor Emissions. EPA/6CJ/6-90/003, Office of Health and Environmental
Assessment, Washington, DC, 1990.
2. U. S. Environmental Protection Agercy, Guideline on Air Quality Models (Hevised)T EPA-450/2-
78-027R, U.S. Environmental Protection Agency, Research Triangle Park, 1987.
8

-------
94-TA23.05
3.	K. S. Rao and L. Satterfield, MPTFR-DS: The MPTER Model Including Deposition and
Sedimentation. EPA/600/8-82/024, U.S. Environmental Protection Agency, Research Triangle Park,
1982.
4.	California Air Resources Board, Subroutines for Calculating Dry Deposition Velocities Using
5.	W. G. N. Slinn, "Precipitation scavenging", in Atmospheric Science and Power Production D.
Randerson, Ed. DOE/TIC-27601, U.S. Department of Energy, Oak Ridge, 1984, pp 466-532.
6.	PEI Associates, Inc. and RE. Cramer Company, Inc., Air Quality Modeling Analysis of
Municipal Waste Combustors. U.S. Environmental Protection Agency, 1986.
7.	J. F. Bowers, J. R Bjorklund, and C.S. Cheney. Industrial Source Complex flSO Dispersion
Model User's Guide. Volume I. EPA/450/4-79/030, U.S. Environmental Protection Agency, Research
Triangle Park, 1979.
8.	U. S. Environmental Protection Agency, User's Guide for the Industrial Source Complex fISC2^
PirpeTsion Moriek Volume II - Description of Model Algorithms. EPA-450/4-92-008b, U.S.
Environmental Protection Agency, Research Triangle Park, 1992.
9.	U. S. Environmental Protection Agency, Development and Testing of Dry Deposition Algorithms.
EPA-454/R-92-GI7, U.S. Environmental Protection Agency, Research Triangle Park, 1993.
10.	U. S. En vironmental Protection Agency, Comparison of a Revised Area Source Algorithm for
ilit Industrial Source Complex Short Term Model With Wind Tunnel Data. EPA-454/R-92-014, U.S.
Environnici.id Protection Agency, Re,c?rch Triangle Park, 1992.
- 11. A. H. Huber and \Y. R Snyder, "Building wahe effects on short stack effluents", in Preprints
Voi-jrne for the Third Symposium on Atmospheric Ditik-.ipn and Air Quality. American
Meteuiolo^cc-il Society, Boston, 1976.
12.	J. S. Scire and L.L. Sch'ilman, "Modeling plume rise from low-level buoyant line and point
sources", in Proceedings Second Je;m Conference on Application of .Air Pollution Meteorology.
American Meteorological Society, jncyv Orleans, 1980, pp 133-139.
13.	K. S. Rao, Analytical Solutions of a Gradient-Transfer Model for Plume Deposition and
Sedimentation. NOAA Technical Memorandum EP.L ARI-109, National Oceanic and Atmospheric
Administration, Silver Springs, 1981.
14.	J. A. A. Venkatram, and R Yamartino, ADOM/TADAP Model Development Program.
Volume 4. Ihe I>.y Deposition Module. Ontario Ministry of the Environment, Rexdale, 1984.
15.	T.W. Horst, "A correction to the Gaussian source-depletion model", in Precipitation Scavenging-
Pry Deposition and Resiii-per^lcn. R R Pruppacher. R G. Semonin, W. G. N. Slinn, Eds . Elsevier,
New Yoik, 1983.
o

-------
94-TA23.05
16.	G. A Sehmel and W. R Hodgson, A Model for Predicting Dry Deposition of Particles and
Gases to Environmental Surfaces. PNL-SA-6721, Battelle Pacific Northwest Laboratory, Richland,
1978.
17.	J. A. Businger, J.C. Wyngaard, Y. Izumi and E.F. Bradley, "Flux-profile relationships in the
atmospheric surface layer", J. Atmos. Sci., 28: 181-189 (1971).
18.	S. A. Slinn and W. G. N. Slinn, "Predictions for particle deposition on natural waters", Atmos.
Environ, 14: 1013-1016 (1980).
19.	ML Wesely and B. B. Hicks, "Some factors that affect the deposition rates of sulfur dioxide
and similar gases on vegetation", J. Air Poll. Control Assoc, 27: 1110-1116 (1977).
20.	G. J. McRae, W. R. Goodin, and J. R Seinfeld, Mathematical Modeling of Photochemical Air
Pollution, Section 4.4. Estimation of the Monin-Ohukhov length. EQL Report No. 18, California
Institute of Technology, Pasadena, 1982.
21.	G. J. McRae, Mathematical Modeling of Photochemical Air Pollution. Chapter 4. Turbulent
Diffusion Coefficients. Ph. D. Thesis, Env. Engr. Sci. Dept., California Institute of Technology,
Pasadena, 1981.
22.	A. A M Holtslag and A P. van Ulden," A simple scheme for daytime estimates of the surface
fluxes from routine weather data", J. Clim. and Appl. Meteor., 22: 517-529 (1983).
23.	I. T. Chang and P. C. Chen, "Estimations of heat and momentum fluxes near the ground",
Proceedings Second Joint Conference on Applications of Air Pollution Meteorology. American
Meteorological Society, Boston, pp 764-769, 1980.
24.	J. C. Doran and T. W. Horst, "An evaluation of Gaussian plume-depletion models with dual-
tracer field measurements", Atmos. Environ., 19: 939-951 (1985).
25.	J. S. Scire and RJ. Yamaitino, Model Formulation and User's Guide for the CAT.PTJFF
Dispersion Model. Sigma Research Corporation, Concord, MA 01742. (1990)
10

-------
94-TA23.05
Table 1. Comparison of micromcteorological variables computed for use by CON1PDEP and ISC-
COMPDEP1.
COMPDFP	1SC-COMPDLP
Stability
Wind Speed
Fru-t'on
Monin-
Friction
Monin-
Class

Velocity
Obukhov
Velocity
Obukhov



Length
Length
6
2.06
0.163
21 7
0.237
45.6
5
3.09
0.362
106.2
0.463
-651.1
4
2.57
0.343
1000.0
0.415
-81.9
3
2.57
0.368
-134.8
0.428
-54.5
2
2.06
0.350
-22.5
0.366
-26.4
1
2.57
0.501
-9.6
0.440
-39.5
1 Oher nuruwrolo^ical panniers wrre set as follows: Zg - 0..*> m. z, ¦- 2.50 m, albaio - 0.12, moisture »\'
-------
94-1^.23.05
O
>-
H
U
s
u:
I-
CO
o
a.
u:
o
COMPDEP
1SC-C0MPDEP
SETTLIVC VELOCITY
10-* 10°	101
PARTICLE DIAMETER (MICRONS)
-igure 1. Deposition velocities and settling velocity predicted by COMPDEP and 1SC-COMPDEP
for neutral stability, u* = 0.1 nvs, Zq = 0.1 m, and a reference height of 10m
r
0.9
E 0B
u
0.7
>•
f-
u 0.6
0
u 0.5
>
1	04
f-
w
o
c.
u
Q
COMPDEP - 1 um
ISC-COMPDEF - 1 jinri
COMPDEP - !0*im
1SC-C0MPDEP - Idvm
0.2
0.0
0 1 0.2 0.3 0.4 0.5 0.6 0.7 0.B 0.9 1.0
u« (m/s)
Figure 2. Deposition velocities predicted by COMPDEP and ISC-COfvlPDtP for particle diameters
of 1 and 10 (im for neutral stability, Zq = 0.1 m, and a reference height of 10m.
12
-------
TECHNICAL REPORT DAT;

1. REPORT NO
EPA/600/A-94/06O
2.
J.JUL
4. TITLE AND SUBTITLE
IMPROVEMENTS IN INDIRECT EXPOSURE ASSESSMENT
MODELING: A MODEL FOR ESTIMATING AIR CONCENTRATIONS
AND DEPOSITION
	-		—
VREPORT DATE
6. PERFORMING ORGANIZATION CODE
7. AUTBQR(S)
D.B. SCHWEDE, arid J.S. SCIRE2
e. FERFORriTNG ORGANIZATION RIPORT HO.
1. PERFORMING ORGANIZATION NAME AMD AC DRESS
SAME AS BLOCK 12
2 SIGMA RESEARCH CORPORATION
196 BAKER AVENUE
CONCORD, MA 01742
10.PROGRAM ELEMEKT NO.
CC1A1E
11. CONTRACT/GRANT KO.
12. SPONSORING AGENCY NAME AND ADDRESS
ATMOSPHERIC RESEARCH ^ND EXPOSURE ASSESSMENT
LABORATORY
OFFICE OF RESEARCH AN.) DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NC 27711
13.TYPE OF REPORT AND PERIOD COVERED
CONFERENCE PAPER
14. SPONSORING AGENCY COTE
EPA/600/9
IS. SUPPLEMENTARY NOTES
It. ABSTRACT
This paper describes the ISC-COMPDEP model and compares it with its
predecessor, the COMPDEP model. COMPDEP was developed by the U.S. Environmental
Protection Agency (EPA) to fulfill the need for a model to estimate concentrations
and dry and wet depos tion for receptors at all terrain heights for use in indirect
exposure assessments. It was developed by combining the methodologies in the
Industrial Source Complex (ISC) and COMPLEX I models with algorithms for modeling
dry and wet deposition. A recent review of COMPDEP by the U.S. EPA identified
several algorithms in the model as potential areas for improvement. The ISC-
COMPDEP model was developed to address these weaknesses. Improvements were made in
the following algorithms: dry deposition of particles, building wake effects,
deposition in complex terrain, and modeling of area sources. A brief discussion of
the major algorithms and a description of the sensitivity of concentration and
deposition estimates to changes in the model algorithms are presented in this
paper. «.¦'
i».
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/ OPEN ENDED T2PMS
c.COSATI
IS. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS 
UNCLASSIFIED
21.NO. OF PACES
20. SECURITY CLASS (This P«««)
UNCLASSIFIED
22. PRICE
-------