Thus, the 100 grams of Hg
continuous and constant throughout the year.  This is
emitted from the hypothetical
probably a reasonable assumption for coal-fired power
source over the year is estimated to result in a deposition to
plants (the largest-emitting source category in the inven-
Lake Superior of 0.08 to 0.16 grams of total mercury (i.e.
tory), but is less appropriate for many other source
mercury in all forms).  Given that the source was 100 g/yr,
categories.  Even for sources that were relatively continu-
0.08 to 0.16 also expresses the percent of the emissions
ous, data for episodes, such as maintenance or upset-
deposited at this location.
related shut downs, were not included in the inventory.
Weather patterns can be highly episodic and significantly
The above is a simplified description of how the transfer
alter source-receptor relationships; these temporal uncer-
coefficients and emissions data are combined in the model
tainties will certainly compromise the accuracy of the
methodology.  In practice, the multiplication of the
estimated concentrations or deposition at a given location
emissions inventory map and the transfer coefficient map is
at any given time.  However, this analysis has been con-
done numerically, for each mercury form emitted by each
ducted over the course of an entire year (and primarily,
source.  This procedure results in an estimate of the
annual estimates have been generated), and this may
atmospheric deposition impact of each source in the
reduce the uncertainty introduced by this variability.
emissions inventory to each of the Great Lakes.
Linking Transfer Coefficients and Emissions Data
Model Results
To complete the modeling activity, it is necessary to
combine the transfer coefficients and the emission invento-
Overall Atmospheric Deposition to the Great Lakes
ries described previously.  This combination will be
The overall model-estimated deposition amount (kg/yr)
demonstrated by means of a simple example.  In the map
and flux (g/km
in Figure 5 for Hg , it can be seen that there is, for example,
-yr) of mercury to each of the Great Lakes is
a region that refers to transfer coefficient values in the
shown in Figures 13 and 14, respectively, for both wet and
range 0.01 to 0.02 (µg total Hg deposited/km -yr)/(g
dry deposition.  It can be seen that both forms of deposi-
emitted/yr).  Suppose there was a mercury source emitting
tion appear to be important.  Lake Michigan is seen to have
100 grams of Hg  somewhere in that region.  What the
the greatest deposition amount, while Lake Erie appears to
estimated transfer coefficient means is that the estimated
have the highest deposition flux.
deposition flux of mercury resulting in Lake Superior from
that source will be:
Geographical Distribution of Atmospheric Deposition
Contributions of Mercury
As mentioned above, the modeling methodology described
herein generates estimates of the contribution of each
100 (grams Hg  emitted/yr)  •  0.01 to 0.02 (µg total
source in the emissions inventory (~106,000 discrete
Hg deposited/km  - yr)(grams Hg  emitted/yr)
records (sources, either point or area) in the applied
United States / Canadian inventory) on each receptor of
1 to 2 (µg total Hg deposited/km  - yr)
interest.  As a way of summarizing these results, Figures 15-
19 show the geographical distributions of mercury source
To get the actual amount of mercury contributed to the
contributions to atmospheric deposition in each of the
entire surface of Lake Superior from this hypothetical
Great Lakes.
source through atmospheric deposition, one would
multiply by the surface area of the lake.
It can be seen that mercury deposition to each of the lakes
arises from throughout the region, and that even distant
deposition amount
sources can contribute significant amounts.  For example,
even sources in Florida appear to be able to contribute
flux • surface area
significant amounts of mercury to each of the Great Lakes.
The geographical region of significant contributions is
1 to 2 (µg total Hg deposited/km  - yr)  •  81,200
somewhat distinct for each lake, as would be expected
(km )
given their different locations and the variations in the
extent of industrialization and urbanization in each basin.
81,200 to 162,400 (µg total Hg deposited/yr)
For example, there are significant contributions to Lake
Superior from regions within approximately 1000 km west
0.08 to 0.16 (grams total Hg deposited/yr)
of the lake, but the relative contribution of this region to
the other lakes is lower, due to both the increased