PROCEEDINGS OF 1995
CANADIAN MERCURY NETWORK WORKSHOP
[INLINE]
MERCURY IN LAKE AND MARINE SEDIMENTS FROM NORTHERN CANADA
W. L. Lockhart, P. Wilkinson, R. Hunt and R. Wagemann
Contaminants Research Section, Fisheries and Oceans Canada, 501 University Crescent, Winnipeg, Manitoba R3T 2N6
Introduction
Many lakes in northern Canada produce fish with mercury contents
higher than those recommended for consumption by Health Canada.
(Health Canada has recommended that mercury in fish sold
commercially not exceed 0.5 ppm (wet wt) and that the mercury not
exceed 0.2 ppm in subsistence settings where people consume a lot of
fish (Health & Welfare Canada, 1978; 1984). There is wide variation
in natural, geological mercury throughout Canada (Painter et al.,
1994). At the same time, there is evidence that mercury has been
increasing in the atmosphere of the northern hemisphere at a rate of
over 1 per cent per year (Slemr & Langer, 1992), and that mercury
inputs to lakes in much of central North America have increased as a
result of atmospheric fallout (Swain et al., 1992). The question in
the North, then, is the extent to which fish in northern lakes are
being influenced by natural mercury and by mercury from atmospheric
pollution.
The only approach we are aware of which might offer insight into the
question of natural vs. atmospherically transported mercury in
northern Canada is the record preserved in natural archives in
integrating media like lake sediments, glacial cores and peat bogs.
In instances where mercury in recent (last half century or so) media
exceed those in older media, we may infer that the increase has been
caused by increased loadings of mercury. This requires the
assumption that mercury moves very little or not at all after it has
been deposited. Many studies of lake sediments have been studied and
most of them from central North America show a basal period with low
mercury before the present century with a rapid increase in mercury
during the present century followed by a very recent decline. This
pattern has been derived as a world summary of the content of heavy
metals in cores by Valette-Silver, 1992. We opted to examine lakes
in northern Canada in efforts to estimate current fluxes of mercury
to them, and to set those inputs in their appropriate historical and
geological settings.
Methods
Lakes from northwestern Ontario north to the high arctic islands and
west to the southern Yukon have been sampled since 1986 (Lockhart et
al., 1993). In addition several cores have been obtained from Hudson
Bay on the cruises of C.S.S. Hudson in 1993 and M.V. Fogo Isle in
1993. In lakes, we have usually collected cores late in the winter
to take advantage of ice as a working platform. Two types of corers
have been used in lakes, one using circular plastic tubes 10 or 16
cm in diameter and a larger stainless steel box corer. For cores in
Hudson Bay and Lake Winnipeg, a large, oceanographic box corer was
used to collect large boxes of sediment and these were cored by
pushing 10-cm plastic core tubes into them on the deck of the ship.
Cores were sliced into either 1-cm or 0.5-cm slices in the field
just after collection and individual slices were sealed into
Whirlpak bags. They were kept at as near as possible to 4oC until
analyzed. Normally slices were analyzed for lead-210 and cesium-137
before proceeding to more extensive analysis. If the lead-210 showed
a down-core profile approximating exponential decay, and the
cesium-137 showed a peak below the surface, the core was accepted as
one worth further analysis. Periods of deposition of those slices
with excess lead-210 were estimated by several methods, sometimes
using mixing models kindly supplied by Dr. John Robbins (NOAA, Ann
Arbor, MI). Mercury in samples of freeze-dried material from each
slice was measured using flameless atomic absorption
spectrophotometry following digestion with aqua regia.
Results and Discussion
The profile of mercury in sediment from Lake 375 from the Experiment
Lakes Area is presented in Figure 1.
[IMAGE]
Figure 1. Mercury in slices of sediment from Lake 375, Experimental
Lakes Area, Northwestern Ontario (Adapted from Lockhart et al.,
1993).
The results from other lakes in northwestern Ontario were similar,
showing lower concentrations of mercury in deep slices than in those
near the sediment-water interface. Three cores have been analyzed
from Trout lake and they showed a remarkable range of values,
although the basic shape showing increases in upper slices was
present in all of them. Further north, lakes on the west shore of
Hudson Bay showed a similar pattern as did two lakes in Cornwallis
Island. Two lakes even further north (Buchanan Lake on Axel Heiberg
Island and Lake Hazen on northern Ellesmere Island) showed no
obvious pattern but both of these cores were extensively mixed as
judged by the lead-210 profiles. In the western NWT and southern
Yukon the increases in mercury in the upper slices were much less
striking.
In each lake we have tried to estimate the tendency of the core site
to focus sediments so that the core represents an accumulation from
more area than its actual area. This has been done using the flux of
lead-210 at the site and comparing it either to the flux estimated
from soil samples within the drainage or to the flux expected based
on the latitude of the site. Using this factor to calculate focusing
factors, the concentrations of mercury in the top and bottom slices
have been converted to estimates of fluxes calculated from the
product of mercury concentration and sedimentation rate. The fluxes
have then been adjusted by the focusing factors and these new
estimates of corrected fluxes have been compared for tops and
bottoms of cores. The flux at the top is taken as a measure of the
total net mercury flux, comprising contributions from natural and
anthropogenic sources, and the flux at the bottom is taken as a
measure of the natural, geological flux. The difference in these is
taken as the anthropogenic contribution. The results of these
calculations are shown in Figure 2. The pattern suggests that
mercury in lakes in northwestern Ontario and in the eastern NWT
originates mainly from anthropogenic sources while that in the
western NWT and Yukon Territory originates mainly from natural
sources.
[IMAGE] Recently the above interpretation of core profiles (and of
other indications that anthropogenic sources exist) has been
challenged (Rasmussen, 1994). It has been suggested that natural
geochemical gradients rather than increases in deposition result in
mercury profiles which show increased mercury in the upper slices.
One case which offers a test of this is that of Clay Lake in
northwestern Ontario. This lake received mercury from a chlor-alkali
plant until 1969 and a core taken in 1971 showed the peak of mercury
at the surface of the sediments. A second core taken in 1978 showed
the peak a few cm deeper and a third core taken in 1995 showed the
peak a few cm deeper yet. This series of cores indicated that the
mercury did not move to the surface in response to post-depositional
processes but rather became buried deeper and deeper. Similar
results from a chlor-alkali plant in eastern Canada were reported by
Smith and Loring (1981).
Conclusions
We conclude that cores can and often do preserve records of changing
inputs of mercury and that a regional pattern of mercury impacts
exists in northern Canada with eastern sites influenced more by
anthropogenic sources of mercury than western sites.
References
1. Health Welfare Canada. 1978. Methylmercury in Canada. Exposure of
Indian and Inuit Residents to Methylmercury in the Canadian
Environment. 200 pgHealth and Welfare Canada, Medical Services
Branch.
2. ____. 1984. Methylmercury in Canada. Exposure of Indian and Inuit
Residents to Methylmercury in the Canadian Environment. 164 pgHealth
and Welfare Canada, Medical Services Branch.
3. Lockhart, W. L., P. Wilkinson, B. N. Billeck, G. J. Brunskill, R.
V. Hunt, and R. Wagemann. 1993. Polycyclic Aromatic Hydrocarbons and
Mercury in Sediments From Two Isolated Lakes in Central and Northern
Canada. Wat. Sci. Tech. 28, no. 8-9: 43-52.
4. Painter, Scott, Eion M. Cameron, Rod Allan, and Jeremy Rouse.
1994. Reconnaissance Geochemistry and Its Environmental Relevance.
J. Geochem. Explor. 51: 213-46.
5. Rasmussen, Pat E. 1994. Current Methods of Estimating Atmospheric
Mercury Fluxes in Remote Areas. Environ. Sci. Technol. 28, no. 13:
2233-41.
6. Slemr, F, and E. Langer. 1992. Increase in Global Atmospheric
Concentrations of Mercury Inferred From Measurements Over the
Atlantic Ocean. Nature 355: 434--437.
7. Smith, John M., and Douglas H. Loring. 1981. Geochronology for
Mercury Pollution in the Sediments of the Saguenay Fjord, Quebec.
Environ. Sci. Technol. 15: 944-51.
8. Swain, Edward B., Daniel R. Engstrom, Mark E. Brigham, Thomas A.
Henning, and Patrick L. Brezonik. 1992. Increasing Rates of
Atmospheric Mercury Deposition in Midcontinental North America.
Science 257: 784-7.
9. Valette-Silver, Nathalie. 1992. Historical Reconstruction of
Contamination Using Sediment Cores: a Review, Technical Memorandum
NOS/ORCA 65. NOAA, Rockville, Maryland.
[INLINE]
back to intro next section
BACK TO
*********************************************************************