Contemporary
and Historical Eolian Depositional Fluxes of Mercury: Archival Records in
Ombrotrophic Bogs and Lake Sediments from Nova Scotia and New Zealand
C.H. Lamborg1,
W.F. Fitzgerald1, A.W.H. Damman2, J.M. Benoit3,
P.H. Balcom1 and D.R. Engstrom4
1Univ. of Connecticut, Dept. of Marine Sciences, Groton, CT 06349 USA
2Kansas State. Univ., Division. Of Biology, Manhattan, KS 66506-4901 USA
3Princeton Univ., Dept. of Geo. Sciences, Princeton, NJ 08544 USA
4Science Museum of Minnesota, St. Croix Watershed Research Station, St.
Croix, MN 55047 USA
email: william.fitzgerald@uconn.edu
Abstract
Using lake sediments and ombrotrophic bog peat from
remote locations, we are reconstructing the atmospheric deposition of mercury
(Hg) over the last ca. 800 years in both hemispheres. Significant
findings include:
The current flux of
Hg from the atmosphere in Nova Scotia was estimated using three independent
methods: lake sediments dated with 210Pb, ombrotrophic peat dated
with Polytrichum and rain collections
(c/o Environment Canada and the Mercury Deposition Network). These three estimates are very similar (ca.
9 µg/m2/y). Therefore, the
contribution of dry deposition to total deposition is not dominant and
statistically indistinguishable from zero (20±50% of total flux) at this
location.
A factor of 3-4x
increase in the deposition of Hg to the lake sediment archives was observed
since the advent of the Industrial Revolution in 5 cores from Nova Scotia. A similar trend is evident in the first core
from New Zealand. Furthermore, this
increase is synchronous with increases in emissions of CO2 from
fossil fuel combustion on a global scale.
No evidence was
found for an enhancement in atmospheric flux as a result of pre-industrial Au
and Ag mining in either hemisphere.
This finding implies that much of the Hg lost during mining operations
ended up in soils and sediments and is now largely immobile with respect to
atmospheric emission.
Introduction
Mercury (Hg) is
released to the atmosphere by natural and human-related processes (e.g., Nriagu
and Pacyna, 1988). In the atmosphere,
Hgº vapor is the dominant chemical form and is slow to oxidize to more soluble
species (e.g., Lamborg et al.,
2000). It is therefore available to be
widely dispersed within the atmosphere, both intra- and interhemispherically
(e.g., Fitzgerald, 1995). This implies
that the atmospheric depositional flux of Hg at any location is integrative of
sources on large and small spatial scales.
Many studies have exploited this global/regional nature of Hg dispersion
to estimate the change in the atmospheric burden of Hg on a global-scale by
examining the depositional flux recorded over time in a natural archive from a
remote location (e.g., Steinnes and
Andersson, 1991; Swain et al., 1992;
Stewart and Fergusson, 1994; Landers
et al., 1995; Norton et al., 1997; Benoit et al., 1998; Lockhart et
al., 1998; Rognerud et al.,
1998; Lacerda et al., 1999; Martínez-Cortizas et
al., 1999; Matsunaga et al.,
1999). Most of these studies have used lake
sediments or ombrotrophic peat as the archiving media and results from these
studies are generally in agreement (Fitzgerald et al., 1998). The picture
that is emerging is one of a widespread increase in Hg deposition since the
Industrial Revolution (ca. 1890 c.e.).
Rarely, however, have these studies been conducted in conjunction with
contemporaneous precipitation collections, and rarer still are studies that
included more than one archive type from the same location. Furthermore, information from the Southern
Hemisphere is much more sparse than for the Northern Hemisphere. We report here findings on the Hg
accumulation rates in lake sediments and ombrotrophic peat from two locations
that are remote and representative of their respective hemispheres.
Methods
Ombrotrophic bogs
and seepage/headwater lakes were sampled in Nova Scotia and New Zealand (Figure
1). To minimize confounding effects
from diagenesis within peat (Damman, 1978), cores were only collected from the
raised, central region of the bogs.
Only the results from bog microtopography raised above the local water
table (hummocks) are reported. To
minimize confounding factors in sediment studies, lakes were selected for
simple morphology, small catchment size and minimal surface water
exchange. Multiple cores were collected
from each bog and lake, and averaged results are reported where possible.
The methods for peat
coring, handling, preparation and analysis for Hg and 210Pb used can
be found in Benoit et al.
(1998). The methods for lake sediment
collection and analysis closely followed those of Swain et al. (1992).
Results
We found that
application of a dating model such as the constant rate of supply (CRS) using 210Pb
as the geochronological tracer was not feasible in the peat sampled from either
Nova Scotia or New Zealand as a result of apparent Pb mobility. Unlike Pb, the Hg profiles as well as
previous biogeochemical assessment of peat as a Hg archive (Benoit et al., 1998) indicated that Hg was not
significantly mobile above the water table.
In Nova Scotia, however, we used Polytrichum
as an alternative dating technique for the first 15 cm (below this depth, the Polytrichum became too decomposed to
reliably record the number of winter-time kinks and therefore the mass
accumulation rate). CRS dating was
feasible in the lake sediments examined in both locations. As part of the Mercury Deposition Network,
the Atmospheric Environment Service of Environment Canada collects rainwater
for Hg analysis at Kejimkujik National Park in Nova Scotia (where our lakes
were situated). We therefore have three
independent estimates of the current flux of Hg to Nova Scotia: precipitation,
surficial lake sediments and peat dated by Polytrichum. These estimates agree closely (rain: 9±3;
lake: 9±2; peat: 11±4 µg/m2/y; Figure 2). The two archiving media might be expected to deviate from the wet
depositional signal if dry depositional mechanisms were of sufficient
magnitude. The similarity of the three
estimates of contemporary fluxes suggests that dry deposition is not the
dominant atmospheric removal mechanism in this location (20±50% of total;
assuming the magnitude of the difference between the rain measurements and the
peat reconstructions is an upper estimate of the difference between the wet
flux and the total flux).
Unfortunately, Polytrichum is
not found in Southern Hemispheric bogs, nor is there a current project to
collect rainwater for Hg determination in the region of New Zealand’s South
Island pertinent to our archive studies.
The temporal change
in Hg deposition is well recorded in lake sediments from both locations (Figure
3). With some of the cores,
reconstruction of almost 800 years of deposition was possible, and most cores
recorded several hundred years. In both
hemispheres, a clear increase in the flux of Hg to the sediments occurs around
the mid-1800’s and current flux estimates appear to be about 3-4x that of the
pre-industrial signals (i.e., an increase of 200-300%). Included in Figure 3 is an estimate of Hg
inputs to the global atmosphere developed by Hudson et al. (1995) based on the work of Nriagu (1994) as well as fossil
fuel consumption estimates contained in the Trends
database (Keeling, 1994; Marland et al.,
1994). The Hg emissions signals
associated with pre-1900 mining activities in North and South America (Au and
Ag, principally) from that analysis are not recorded in either location. However, the advent and time rate of change
for fossil fuel combustion associated with the Industrial Revolution fits well
with the character of the last hundred years of Hg deposition. The lack of a mining signal in lake
sediments from the Southern Hemisphere has also recently been reported by
Lacerda et al. (1999) working in
Brazil and previously noted by several authors working in locations around the
Northern Hemisphere (e.g., Swain et al., 1992; Landers et al.,
1995; Lockhart et al., 1998; Rognerud et al., 1998; Martínez-Cortizas
et al., 1999; Matsunaga et al., 1999). While the losses of Hg Au mining is fairly
well documented (Nriagu, 1994), the lack of a global Hg signal associated with
mining suggests that either the Hg emitted to the atmosphere from these
activities is removed rapidly and locally, or perhaps more reasonably, that
most of this Hg was lost largely to soils and sediments and is no longer
globally active on the time scale of these reconstructions.
Acknowledgements
We thank the
following for their indispensable help: Connie Langer, David Cohen, Steve
Beauchamp, Rob Tordon, Bob Thexton, Chris Waterman, Jonathon Kim, Paul
Meredith, Keith Hunter, Ming Chang, Amit Dave, Kelley Thommes, Fjordland
National Park (NZ) and Kejimkujik National Park (NS), Cooper’s Inn, Whitman
Inn, Gary Grenier, Bob Dziomba, Don Porcella and Mary Ann Allen. This research was supported financially by
the Electric Power Research Institute.
This is contribution xxx from the UConn Marine Science and Technology
Center.
References
Benoit J.M., Fitzgerald W.F. and Damman
A.W.H. (1998) Environ. Research 78: 118-133.
Damman A.W.H. (1978) Oikos 30: 480-495.
Fitzgerald
W. F., Engstrom D.R., Mason R.P. and Nater E. A. (1998) Environ. Sci. And Technol. 32:
1-7.
Fitzgerald
W.F. (1995) Water Air and Soil Pollution
80: 245-254.
Hudson R.J.M., Gherini S.A., Fitzgerald W.F.
and Porcella D.B. (1995) Water Air Soil
Poll. 80: 265-272.
Keeling C.D. (1994)
In: Trends ’93: A Compendium of Data on
Global Change. T.A. Boden, D.P. Kaiser, R.J. Stepanski and F.W. Stoss
(eds.) ORNL/CDIAC-65, Oak Ridge National Lab., U.S.A. pp. 501-504.
Lacerda
L.D.; Ribeiro Jr. M.G.; Cordeiro R.C.; Turcq B. and Sifeddine A. (1999) Mercury
Atmospheric Deposition to Northern South America During the Last 1,000 Years.
Presented at: 5th
International Conference on Mercury as a Global Pollutant, Rio de Janeiro.
Lamborg
C.H., Fitzgerald W.F., Graustein W.C. and Turekian K.K. (2000) J. Atmos. Chem. 36/3: 325-338.
Landers
D.H., Ford J., Gubala C., Monetti M., Lasorsa B.K., Martinson J. (1995) Water Air Soil Poll. 80/1-4: 591-601.
Lockhart
W.L., Wilkinson P., Billeck B.N., Danell R.A., Hunt R.V., Brunskill G.J., Delaronde
J. and St Louis V.(1998) Biogeochemistry 40/2-3: 163-173.
Marland G., Andres
R.J. and Boden T.A. (1994) In: Trends
’93: A Compendium of Data on Global Change. T.A. Boden, D.P. Kaiser, R.J.
Stepanski and F.W. Stoss (eds.) ORNL/CDIAC-65, Oak Ridge National Lab., U.S.A.
pp.505-584.
Martínez-Cortizas A., Potevedra-Pombal X.,
García-Rodeja E., Nóvoa Muñoz J.C. and Shotyk W. (1999) Science 284: 939-942.
Matsunaga
T., Ueno T., Chandradjith R.L.R., Amano H., Okumura M., Hashitani H. (1999) Chemosphere 39/2: 269-283.
Norton S.A., Evans G.C. and Kahl J.S. (1997)
Water Air Soil Poll.100: 271-286.
Nriagu
J.O. (1994) Sci. Tot. Env. 149/3: 167-181.
Nriagu J.O. and
Pacyna J.M. (1988) Nature 333: 134-139.
Steinnes E. and Andersson E.M. (1991) Water, Air, Soil Pollut. 56
391-404.
Stewart C. and Fergusson J.E. (1994) Environ. Poll. 86: 243-249.
Swain
E.B., Engstrom D.R., Brigham M.E., Henning T.A. and Brezonik P.L. (1992) Science 257: 784-787.
Figure
1. Sampling locations in
Nova Scotia and New Zealand’s South Island.
Figure 2. Current Hg
deposition estimated from direct measurement, lake sediments and peat in Nova
Scotia
Figure 3. Temporal
variation in Hg flux to lake sediments in N.S. and N.Z.
Estimates of global-scale atmospheric emissions from
mining and general industry taken from Hudson et al., 1995.