EPA/600/A-92/234
92-84.17
To be presented at the 85th Annua] Meeting
of the Air & Waste Management Association,
Kansas City, MO, June 23-27, 1992
Model Calculations of the Annual Atmospheric
Deposition of Toxic Metals to Lake Michigan
Terry L. Clark'
Atmospheric Sciences Modeling Division
Air Resources Laboratory
National Oceanic and Atmospheric Administration
Research Triangle Park, North Carolina 27711
Pamela Blakley
Air and Radiation Division
U.S. Environmental Protection Agency
77 West Jackson Street
Chicago, Illinois 60604
George Mapp
Computer Sciences Corporation
3210 Chapel Hill-Nelson Highway
Research Triangle Park, North Carolina 27711
On assignment to the Atmospheric Research and Exposure Assessment Laboratory, U.S.
Environmental Protection Agency

-------
92-84.17
INTRODUCTION
The presence of toxic substances in the Great Lakes creates significant environmental
risks for both human and wildlife populations. Originally, concern was focused on water
quality degradation due to point source discharges into waterways as the primary source of
toxic input. However, for many toxic substances the atmosphere is considered to be a
substantial contributor of loadings.
The 1990 Clean Air Act Amendments include specific provisions to study significant
sources of atmospheric deposition of toxic substances and their impacts on the health and
welfare of the Great Lakes and other major water bodies. This act also requires an
assessment of the atmospheric loadings to the Great Lakes and other major water bodies.
However, because of cost and technological limitations, spatially-integrated atmospheric
deposition to a body of water can not be directly determined. Until recently, the total annual
atmospheric deposition was estimated from the product of (1) spatially-limited, land-based,
rural air concentrations, (2) constant theoretical or empirical atmospheric removal rates, and
(3) the surface area of the water body1.
Although it is expedient, this approach, with its inherent assumptions, has several
serious drawbacks that cast doubts on the spatially-integrated atmospheric loadings. First,
this approach assumes that both air concentrations and removal rates are constant everywhere
over the water body. Secondly, the approach assumes that the land-based rural air
concentrations at one or several remote sites are representative of those over the entire water
body.
These assumptions may not be valid. In reality air concentrations and removal rates
across large areas the size of the Great Lakes are both spatially and temporally
inhomogeneous. Moreover, since the sources of airborne toxic emissions of many pollutants
are concentrated within or near large urban areas, air concentrations near urban areas are
greater than in remote areas. This must be considered when calculating total atmospheric
deposition to the lake. Finally, because of the frequent shallow tropospheric marine layer
and its resulting atmospheric stability, air pollutants may never reach the water surface2.
Thus, the air concentrations at the air/water interface may be considerably lower than those
at the nearby air/land interface.
Comprehensive atmospheric deposition models based on future enhanced versions of
currently-available regional air pollution models offer a more accurate alternative to this
traditional approach. However, since a better understanding of the emission rates and
atmospheric processes governing airborne toxic deposition is a prerequisite for the enhanced
model versions, these comprehensive models will not be operational within the next several
years. In the interim, the Regional Lagrangian Model of Air Pollution (RELMAP)3, a
2

-------
92-84.17
simple atmospheric deposition model, was applied to quantify the expected range of annual
deposition amounts of toxic trace metals to Lake Michigan. These metals are arsenic (As),
cadmium (Cd), chromium (Cr), lead (Pb), and nickel (Ni).
INTERIM MODELING APPROACH
RELMAP is a four-layer Lagrangian puff model that calculates, for each 3-hour time
step, the pollutant mass within and the amount of pollutant mass deposited from each of a
series of puffs. The heights of the top 3 layers are 200 m, 700 m and the monthly mean
height of the daytime mixed layer. Only at night a fourth surface-based layer exists, the
depth of which ranges from 30 m to 50 m, depending on the time of year. Pollutant puffs
are released within the 200-700-m layer every 3 hours from virtual sources located at the
emission-weighted center of each unit-degree source cell (Figure 1).
The model input data consist of hourly precipitation amounts from approximately
2000 sites, surface and 850-mb wind velocities, climatological atmospheric stability
categories and mixing heights, and spatial distributions of 11 land-use categories. As puffs
are transported across the model domain, both dry and wet deposition amounts of As, Cd,
Cr, Pb, and Ni are calculated for the land and water areas of each of the receptor cells
depicted by Figure 1. The model was applied for the entire year of 1985, the year when
U.S. and Canadian emissions data were available.
Dry deposition of particles is calculated for each 3-h time step as the product of a
surface-layer air concentration (CJ and a time-dependent dry deposition velocity (vj. Since
this latter term varies with season, land-use, and atmospheric stability, the simulated dry
deposition rates exhibit considerable diurnal, seasonal and spatial variabilities. Currently the
model assumes that the water surface is always in a liquid phase. Since it also assumes a
uniform particle size, the model was applied twice using two different particle sizes: 0.5
microns and 5.0 microns, the range of particle sizes of importance to regional deposition.
Wet deposition of particles is based on the product of a dimensionless washout ratio and the
hourly precipitation rate (mm/h) raised to the power of 0.6224.
In this study annual air emissions of the toxic metals were calculated from point and
area source parameters in the 1985 National Acid Precipitation Assessment Program
(NAPAP) emissions inventory and published pollutant emission factors. Emissions from both
Canadian and United States sources were then gridded to the unit-degree RELMAP
configuration. Although emissions from all states and provinces within the Figure 1 domain
were used in the model applications, Table 1 provides the annual emission rates for each
state within Region V, the states nearest Lake Michigan. By virtue of the proximity of the
sources to the lake, emissions from these states will have a significant contribution to the
deposition to Lake Michigan.
3

-------
92-84.17
MODEL CALCULATIONS OF ANNUAL DEPOSITION
The annual deposition amounts of any pollutant to Lake Michigan depend on the
emission rate, proximity of sources, meteorological factors, and removal efficiency. Since
the emissions rate of Pb was an order of magnitude greater than the emission rates of the
other four metals, its annual deposition to Lake Michigan was also an order of magnitude
greater, as Table 2 demonstrates. Pb deposition, approaching 700,000 kg/yr, dominated the
total deposition of the remaining metals, 18,000 kg/yr to 58,000 kg/yr.
Table 2 also shows that the relative contribution of dry deposition to total deposition
is highly dependent on the assumed particle diameter. For smaller particle sizes (e.g., 0.5
micron) dry deposition accounted for 10% or less of the total deposition. On the other hand,
dry deposition from larger particles (e.g., 5.0 microns) accounted for nearly 40% of the total
deposition. The total deposition for the two particle sizes, however, differs by less than 20%
for As and by approximately 10% for the other metals.
The slight difference in total deposition can be explained by the fact that when dry
deposition is small more mass exists in the atmosphere for wet deposition. The greater
difference in total As deposition amounts appears to be linked to the spatial distribution of As
sources; As emissions from sources near the lake were much lower than the emissions of the
other metals. That is, long range transport appears to play a more important role for As.
An independent estimate of atmospheric Pb deposition to Lake Michigan was
presented by Strachan and Eisenreich1. They based their annual estimate on a typical
precipitation rate, washout rate, dry deposition velocity, and air concentration. From this
mass balance exercise, they estimated an annual atmospheric deposition rate of 543,000
kg/yr, approximately 20% lower than the RELMAP calculation. It is not suiprising for the
model calculation to be greater than the mass balance estimate, since the latter approach
ignores the urban contribution the former approach considers. That is, typical air
concentrations used in the mass balance approach reflect only the lower rural/remote air
concentrations only.
CONCLUSIONS AND FUTURE PLANS
Although toxic emission inventories and current regional toxic models are in states of
infancy, annual atmospheric deposition to Lake Michigan was calculated to provide a
preliminary assessment. As the quality and completeness of emission inventories improve
and as modeled processes are refined, these annual deposition amounts will be recalculated
and uncertainties will be reduced.
4

-------
92-84.17
This preliminary modeling exercise for five toxic metals indicated that Cd deposition
was lowest of the five toxic metals, approximately 20,000 kg/yr. Cr and Ni deposition
ranged approximately from 26,000 kg/yr to 35,000 kg/yr. As, showing a wide range of
deposition amounts based on the range of assumed particle sizes, varied from approximately
40,000 kg/yr to 58,000 kg/yr. Pb deposition by far was the greatest, ranging from 675,000
kg/yr to nearly 690,000 kg/yr. The 1985-calculated Pb deposition amount is likely greater
than that for subsequent years, since Pb emissions from motor vehicles decreased
dramatically after the mid-1980's.
In the future, using source-classification-code (SCC) specific emission grids for each
pollutant, RELMAP will determine the relative contribution of various source types to the
total deposition to Lake Michigan. The modeling exercise will also be extended to Lakes
Superior, Huron, Erie, and Ontario, in that order. Efforts will continue to compare
published monitoring values and deposition estimates with modeled air concentrations and
deposition amounts.
A C KNO VVLEDG EMENTS
The authors wish to recognize the valuable air toxics inventory efforts of Bill Benjey,
Dale Coventry, and Alfrieda Rankins of the Atmospheric Research and Exposure Assessment
Laboratory at Research Triangle Park, North Carolina. Without their assistance this
modeling effort would not have been possible.
NOTICE
This paper has been reviewed in accordance with U.S. Environmental Protection
Agency's peer review and administrative review policies and approved for presentation and
publication. Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
5

-------
92-84.17
REFERENCES
1.	W.M.J. Strachan and S J. Eisenreich, Mass Balancing of Toxic Chemicals in the
Great Lakes: The Role of Atmospheric Deposition, Report to the International Joint
Commission, Windsor, Ontario, 1988, pp. 50-52.
2.	W.A. Lyons, R.A. Pielke, J.L. Eastman, et al, "Mesoscale numerical model
evaluations of the meteorological factors associated with elevated ozone levels in the
southern Lake Michigan region", in Proceedings of the 1991 AMS/AWMA
Conference on Applications of Air Pollution Meteorology, P7.10, American
Meteorological Society, Boston, MA, 1991, pp. 164-167.
3.	B.K. Eder, D.H. Coventry, T.L. Clark, et al, BELMAP: A RegionalLagrangian
Model of Air Pollution User's Guide, EPA-600/8-86/013, U.S. Environmental
Protection Agency, Research Triangle Park, NC, 1986.
4.	T.L. Clark, R.L. Dennis, E.C. Voldner, et al., The International Sulfur Deposition
Model Evaluation, EPA/600/3-87/008, U.S. Environmental Protection Agency,
Research Triangle Park, NC, 1987, p. 20.
6

-------
92-84.17
TABLE 1. Annual emissions of
toxic metals from states within REGION V
Based on 1985 NAPAP Emissions Data
and Emission/Speciation Factors from Various Years
Pollutant Emissions Rate State
(kg/yr)
Arsenic	1976	Total
798
653
205
141
106
73
Michigan
Wisconsin
Minnesota
Illinois
Ohio
Indiana
Cadmium
656
234
188
79
69
61
25
Total
Minnesota
Illinois
Ohio
Indiana
Michigan
Wisconsin
Chromium 1183
Total
Indiana
Illinois
Michigan
Ohio
Minnesota
Wisconsin
318
275
234
219
99
38
7

-------
92-84.17
TABLE 1. Continued
Pollutant
Emissions Rate
(kg/yr)
State
Lead
28518
6287
6174
5182
4111
4084
2680
Total
Illinois
Ohio
Michigan
Indiana
Minnesota
Wisconsin
Nickel
983
384
154
141
122
115
66
Total
Illinois
Indiana
Ohio
Michigan
Minnesota
Wisconsin
8

-------
92-84.17
TABLE 2. RELMAP calculations of the 1985
deposition (kg/yr) of toxic metals to Lake Michigan
Pollutant
Particle
Size
(jim)
Dry
Deposition
Wet
Deposition
Total
Arsenic
0.5
5,870 (10.0%)
52,552 (90.0%)
58,422

5.0
19,033 (39.2%)
29,484 (60.8%)
48,517
Cadmium
0.5
2,053 (10.0%)
18,575 (90.0%)
20,628

5.0
7,179 (39.2%)
11,122 (60.8%)
18,301
Chromium
0.5
2,908 ( 8.7%)
30,678 (91.3%)
33,586

5.0
13,129 (37.1%)
22,275 (62.9%)
35,404
Lead
0.5
62,456 ( 9.1%)
627,306 (90.9%)
689,762

5.0
254,201 (37.7%)
420,818 (62.3%)
675,019
Nickel
0.5
2,483 ( 8.3%)
27,489 (91.7%)
29,972

5.0
9,539 (35.9%)
17,069 (64.1%)
26,608
9

-------
92-84.17
FIGURE 1. The unit-degree grid configuration of the Regional
Lagrangian Model of Air Pollution (RELMAP).
10

-------
TECHNICAL REPORT DATA
(Phase read Instruction! on in e reverse before complet
1, REPORT NO, 12.
EPA/600/A-92/234 J
3.
4, title and subtitle
MODEL CALCULATIONS OF THE ANNUAL ATMOSPHERIC
DEPOSITION OF TOXIC METALS TO LAKE MICHIGAN
5, REPORT DATE
6. PERFORMING ORGANIZATION CODE
7, AUTHOR(S)
Terry L. Clark
a. performing organization report no.
9. PERFORMING ORGANIZATION NAME and address
Same as 12.
10. PROGRAM ELEMENT NO.
A101G/C/72/01
11. contract/grant no,
Inhouse
12. SPONSORING AGENCY NAME AND ADDRESS
Atmospheric Research & Exposure Assessment Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
13. TYPE OP REPORT AND PERIOD COVERED
Interim 4/91-5/92
14. SPONSORING AGENCY CODE
EPA/600/09
15, SUPPLEMENTARY NOTES
Presented at the 85th Annual Meetino of the Air & Waste Management Association,
Kansas Htv. MH., .liinp 23-27, 1992
16, ABSTRACT
_^^>Tbe 1990 Clean Air Act Amendment include specific provisions to study significant sources of
* atmospheric deposition of toxic substances and their impacts on the health and welfare of the Great Lakes and
other major water bodies. This act also requires an assessment of the atmospheric loadings to the Great Lakes
and other major water bodies. However, because of cost and technological limitations, spatially-integrated
atmospheric deposition to a body of water can not be directly determined. Until recently, the total annual
atmospheric deposition was estimated from the product of (1) spatially-limited, land-based, rural air
concentrations, (2) constant theoretical or empirical atmospheric removal rates, and (3) the surface area of the
water body. Although it is expedient, this approach, with its inherent assumptions, has several serious
drawbacks that cast doubts on the spatially-integrated atmospheric loadings.
Comprehensive atmospheric deposition models based on future enhanced versions of currently-available
regional air pollution models offer a more accurate alternative to this traditional approach. However, since a
better understanding of the emission rates and atmospheric processes governing airborne toxic deposition is a
prerequisite for the enhanced model versions, these comprehensive models will not be operational within the
next several years. In the interim, the Regional Lagrangian Model of Air Pollution (RELMAP), a
simple atmospheric deposition mode], was applied to quantify the expected range of annual deposition amounts
of toxic trace metals to Lake Michigan. These metals are arsenic (As), cadmium (Cd), chromium (Cr), lead
(Pb), and nickel (Ni). <5===.
17. KEY WORDS AND DOCUMENT ANALYSIS
a, DESCRIPTORS
b.i-ENTiFIERS.'CBEN ENDED TERMS
c. COS AT i Field,'Croup
•V.


18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19, SECURITY CLASS (Tins RtpO'tl
UNCLASSIFIED
21. NO. OP PAGES
11
20, SECURITY CLASS (This past)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (R«v.'4_77) PREVIOUS COITtON l» OBSOL.CTC

-------