United States
Environmental Protection
Agency
Hazardous Waste Engineering
Research Laboratory
Cincinnati OH 45268
*
Research and Development
EPA/600/S2-87/071 Nov. 1987
Project Summary
Sensitivity Analysis for Application
of the Inhalation Exposure
Methodology (IEM) to Studies of
Hazardous Waste Management
Facilities
F. R. O'Donnell and C. C. Gilmore
This study investigated the uncer-
tainties associated with using the
Inhalation Exposure Methodology (IEM)
to determine human exposures to
hazardous waste management facility
air emissions. The Inhalation Exposure
Methodology is an integrated system of
computer programs that simulates the
atmospheric transport of and the result-
ing human exposures to pollutants
released from one or more sources at
an industrial complex. The full report
illustrates the sensitivity of IEM pre-
dictions to (1) variations of important
user-supplied source, meteorological,
and pollutant parameter values and (2)
use of three IEM source modeling
options to represent emission sources
found at hazardous waste management
facilities.
This Project Summary was developed
by EPA's Hazardous Waste Engineering
Research Laboratory, Cincinnati, OH, to
announce key findings of the research
project that Is fully documented In a
separate report of the same title (see
Project Report ordering Information at
back).
Introduction
The Inhalation Exposure Methodology
(IEM) is an integrated system of computer
programs that simulates the atmospheric
transport of and the resulting human
exposures to pollutants released from
one or more sources at an industrial
complex. This study was undertaken to
determine the sensitivity of IEM predic-
tions of pollutant concentrations and
population exposures to (1) variations of
selected, user-supplied source, meteoro-
logical, climatological, and pollutant pa-
rameter values and (2) use of the three
available source modeling options to
represent emission sources found at
hazardous waste management facilities
(HWMFs). These sources include incin-
erators and associated structures, storage
and treatment tanks, drum stacks, process
buildings, surface impoundments, land-
fills, waste piles, and land treatment
areas. Several sources may be found at
one HWMF.
The study only determined the sen-
sitivity of IEM predictions to the above
factors. It did not validate the model by
comparing IEM predictions with actual
field data.
Modeling the sources found at an
HWMF could present problems because
they may be located close together or
near buildings and structures that could
influence pollutant dispersion, and they
may have ill-defined pollutant release
rates. In some cases, source-specific pol-
lutant release rates may be unavailable,
thus forcing the modeler to represent
several sources at a single source.
The IEM uses a Garissian-plume
atmospheric dispersion model, the Inhala-
tion Source Complex Long Term Model
(ISCLTM), to calculate annual-average,
sector-averaged, centerline, ground-level,
air concentrations of released pollutants
at user-selected receptor points. It uses
these concentrations to calculate average
-------
concentrations over each sector segment
of a user-specified polar grid. Finally, it
multiplies the sector-segment-averaged
concentrations and their corresponding
sector-segment populations to give esti-
mates of human exposures to the released
pollutants. Although applicable to a
variety of problems, the IEM was devel-
oped as a tool for estimating pollutant
concentrations and associated human
exposures in the vicinity of hazardous
waste management facilities (HWMFs).
Approach
Emission sources found at HWMFs have
relatively low pollutant release heights,
may be located near structures that in-
fluence pollutant dispersion, and, except
for incinerator stacks, may have es-
sentially no associated plume rise. Pre-
vious studies have examined the sen-
sitivity of ISCLTM predictions to typical
hazardous waste incinerator stack pa-
rameters (e.g. stack height, gas tempera-
tures). Based on these studies and the
fact that all stack parameters except the
physical stack height affect only plume
rise, these parameters were not studied
in detail. The remaining, important, user-
supplied input parameters include mete-
orological parameters (wind speed, wind
direction, and stability class), source
parameters (release height, source area,
and adjacent building cross sections),
pollutant parameters (decay coefficient,
settling velocity, and reflection coeffici-
ent), and the array of receptor grid points
chosen.
The study report documents the effects
of varying these parameters on ambient
pollution concentrations and population
exposures. In addition, the effects of using
three different source representation
options (point, area or volume representa-
tions) on pollution concentrations are
investigated.
Several typical HWMF sources were
selected for detailed study; a stack with
essentially no plume rise, a 14.1 -m square
(200 m2) area, an 80.6-m square (6500
m2) area, a 316.2-m square (100,000 m2)
area, and a 2236.1 -m square (5,000,000
m2) area. Since the ISCLTM algorithm
will not accept zero values for a stack
diameter or gas exit velocity, our stack
source was assumed to have a diameter
of 1.0 m and a gas exit velocity of 1 x 10 5
m/s, the ISCLTM default value. Source
(release) heights of 0,5,10,15, and 20 m
were considered for these sources.
Limited evaluations were made of the
effects of representing a 200-m2 process
building with a release hefght of either 5
or 10 m by one stack source, by two area
sources, and by two volume sources.
Similar evaluations were made for a
200-m2 tank farm containing four tanks
with release heights of either 3 or 6 m
that were represented by four point
sources, with and without building wake
effects; by ^ne area source; by four area
sources; by one volume source; and by
four volume sources.
In order to investigate the sensitivity of
different IEM input parameters and pro-
gramming options, the following computer
outputs were generated:
1. Plots of the exceptor grid-point con-
centrations directly downwind of the
source. (These are defined as pri-
mary grid-point concentrations).
2. Value and location of the maximum
primary grid-point concentration.
3. The exposure to all individuals living
directly downwind of the source.
(The area directly downwind of the
source is defined as the "primary
sector," which lies within ± 11.25°
of the wind direction.)
4. Plots of the sum of the pollution
concentrations for all grid points at
a given distance from the source,
which indicates total exposure as a
function of distance from the source.
5. The magnitude and distance from
the source of the maximum con-
centration for each profile generated
in Item 4.
6. The total potential exposure to the
population based on summing the
exposure potentials over all direc-
tions and distances from the source
out to 50Km.
The analyses specified in Items 4-6
were included because the dimensions
of some of the area sources were large
enough to cause substantial air concen-
trations to occur at grid points that lie
outside the primary sector. Ignoring these
concentrations would give a false impres-
sion of the importance of area size. These
measures also give a better picture of
IEM predictions under real meteorological
conditions.
Differences in concentration and ex-
posure potential predictions due to the
choice of source representation option
were also investigated. Two typical HWMF
sources were chosen, a process building
and a small tank farm.
The process building was assumed to
be 10-m high, to cover 2 m2, and to
release pollutants either from a rooftop
or a midheight vent. The building was
modeled as one stack source, as one
14.1 -m square area source, as two 10-m
square area sources, and as two volume
sources having standard deviations of
2.33 for their crosswind source distribu-
tions and 4.65 for their vertical source
distributions.
The tank farm was assumed to contain
four 6.1-m high tanks, to cover 200 m2,
and to release pollutants from vents
located on top of the tanks. The tanks
were modeled as four stack sources, as
four stack sources with adjacent 6.0-m2
high structures, as one 14.1-m square
area source, as four 7.07-m square area
sources, as one volume source having
standard deviations of 3.29 for its cross-
wind source distribution and 2.84 for its
vertical source distribution, and as four
volume sources having standard devia-
tions of 1.64 for their crosswind source
distributions and 2.84 for their vertical
source distributions. Midheight (3.05-m)
releases were considered only for the
single area and volume source repre-
sentations.
Results and Conclusions
Based on the analysis of variations in
user-supplied input parameters and of
the use of several modeling options for
representing emissions sources, the study
made the following findings:
1. Predicted ground-level air concen-
trations are probably accurate to
within a factor of 2, if the IEM is
applied under well-behaved mete-
orological conditions over flat
terrain.
2. The IEM method for estimating the
total exposed population is as ac-
curate as any other general method.
However, the accuracy of the
method used to link exposed per-
sons to specific pollutant concen-
trations (i.e., to calculate exposures)
is unknown, but likely is compar-
able to the accuracy of other exist-
ing methods.
3. For the sources considered in this
study, wind speed acted as a linear
scaling factor, except when pol-
lutant decay and decomposition
were considered. This relationship
also would not hold for stacks that
have an associated plume rise.
4. The effects on predicted pollutant
concentrations due to variations in
atmospheric stability, pollutant
release height, and source area
are interdependent. All three pa-
rameters are strongly influenced
by predicted concentrations, and
every effort should be made to use
accurate values for them.
-------
5. Increasing atmospheric stability
increased exposure estimates, but
it may either increase or decrease
maximum concentration predic-
tions, depending largely on the
release height.
6. Increasing release height de-
creased both exposure and con-
centration estimates.
7. Increasing source area had little
effect on exposure estimates for
the same receptor array. Maximum
concentration predictions varied by
as much as 60% for the source
areas considered in this study.
8. Use of the building wake effects
option increased concentration
predictions within 200 m of the
source center, but had little effect
on more distant concentration pre-
dictions and on exposure estimates.
9. For pollutants that have half-lives
of a few days or less, pollutant
decay could significantly reduce
airborne concentrations at recep-
tors beyond 1 km. For longer-lived
pollutants, decay is unimportant.
10. Pollutant disposition significantly
affected both concentration and
exposure predictions, especially at
sites characterized by stable atmo-
spheric conditions and low wind
speeds. The pollutant deposition
option in IEM should be used if the
emitted pollutants are particles or
can form particles that can be
characterized.
11. The choice of a receptor array can
bias predictions significantly. An
array with receptors concentrated
between the minimum allowed-
radial receptor distance and 2 km
should produce the most accurate
estimates of maximum concentra-
tions and exposures.
12. The various available emissions
source modeling options produced
essentially the same exposure esti-
mates and airborne concentrations
at receptors beyond approximately
one kilometer. At the closer recep-
tors, the stack and the area source
representations produced very
similar results. Volume source
representations predicted close-in
concentrations higher than those
predicted using stack and area
source representations for the more
stable atmospheric conditions. For
the less stable conditions, volume
sources tended to predict the
close-in concentrations which were
lower than for the other two
options.
References
1. O'Donnell, F. R., P. M. Mason, J. E.
Pierce, G. A. Holton, and E. Dixon,
User's Guide for the Automated
Inhalation Exposure Methodology
(IEM), EPA-600/2-83-029(1983).
F. R. O'Donnell and C. C. Gilmore are with Oak Ridge National Laboratory,
Oak Ridge. TN 37830.
Benjamin L Blaney is the EPA Project Officer (see below).
The complete report, entitled "Sensitivity Analysis for Application of the
Inhalation Exposure Methodology (IEM) to Studies of Hazardous Waste
Management Facilities, "(Order No. PB87-232 641/AS; Cost: $18.95, subject
to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Hazardous Waste Engineering Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
United States
Environmental Protection
Agency
Official Business
Penalty for Private Use $300
EPA/600/S2-87/071
Center for Environmental Research
Information
Cincinnati OH 45268
CICAGO
------- |