PROCEEDINGS OF 1995
CANADIAN MERCURY NETWORK WORKSHOP
[INLINE]
PROCESSES EFFECTING MERCURY AND ASSOCIATED METALS IN LAKE SEDIMENT COLUMNS
W. B. Coker
Geological Survey of Canada, 601 Booth Street, Ottawa, Ontario. K1A 0E8
The questions to be addressed are: "What are the causes of and/or
controls on the distribution of minor and trace elements, including
the observed surface enrichment, in lake sediment columns"; and, in
the context of the first question "What is the relative significance
of the effects of long range atmospheric anthropogenic inputs in
comparison to local geologically controlled inputs and
post-depositional physical, chemical and biological/bacterial
processes on minor and trace element distributions in lake sediment
columns".
Lakes comprise a complex system involving interplay between various
physical, chemical and biological factors (Hutchinson, 1957;
Ruttner, 1963; Mortimer, 1941, 1942 and 1971; Sain and Neufield,
1975; and, Coker et al., 1979). The distribution of heat, and of
suspended and dissolved substances, including gases absorbed from
the atmosphere or generated from sediment degassing, as well as
compositional-textural variation and biological-bacterial activity
within the sediment, are factors which can effect the dispersion of
trace elements in the waters, and accumulation or mobilization of
trace elements in the sediments of lakes.
The literature, as reviewed by Rasmussen (1994), indicates that Hg
can migrate in a variety of inorganic and organic species. In the
reducing environment of deep lake sediments, it is feasible that Hg
may be remobilized as organo-Hg2+ complexes, as Hg2+-organosulfide
complexes, and as dissolved or vapour phase Hgo. Redox conditions
and the stability of organo-Hg complexes are important influences on
the mobility of Hg that should be considered in the interpretation
of surface enrichment. Iron and manganese oxides and hydroxides
precipitate and are stable in surface sediments due to the higher Eh
of surface sediments relative to deep sediments. An enrichment in
iron and manganese near the sediment-water interface is likely to be
accompanied by a decrease in Hg mobility, due to the tendency of Hg
and Hg-bound organic complexes to adsorb onto oxide and hydroxide
surfaces. Second, the breakdown of organic material during
diagenesis appears to be accompanied by a reduction in the integrity
of the organo-Hg association with depth.
Trace metals derived from either natural geological or cultural
sources can accumulate in the upper 5 to 20 cm of lake sediment by
biological and geochemical mechanisms (Mortimer, 1942 and 1971;
Gorham and Swain, 1965; Mackereth, 1966; Cline and Upchurch, 1973;
Farmer, 1991; Rasmussen, 1994). Upward migration of minor and trace
metals may occur because of dewatering due to compaction and
unidirectional ion migration, but, to a much greater extent,
migration appears to be due to bacterial activity (Cline and
Upchurch, 1973). Cline and Upchurch (1973) experimentally
demonstrated that an initial surface enrichment of copper in a lake
sediment column was developed after one week and a well established
enrichment within four weeks. They speculated that the natural
migration of heavy metals was primarily due to a bacterial mechanism
combined with metal transport on bubble interfaces, in a gaseous
complex or as soluble organic complexes, and that stabilization
results from the metal being immobilized as a new complex or as an
inorganic precipitate in the biologically active portion of the
sediment close to the sediment-water interface. Walters et al.
(1972), based on their study of mercury in Lake Erie sediments,
suspected that mercury diffuses upward through the sediment,
resulting in higher concentrations in the sediment-water interface
zone.
In lakes having aquatic flora and fauna as a significant source of
organic matter, or in which waters are highly productive and
sedimentation rate is rapid, the influence of organisms on trace
element distribution may be significant. In the flat-lying,
tree-covered terrain characteristic of the southern Canadian Shield,
and in the terrain of the North American Appalachia, the incidence
of organic matter is high and metal-organic interactions are
predominant. The presence of organic matter can either enhance trace
element mobility, by forming mobile soluble organic complexes or
retard it, by direct precipitation of insoluble organic complexes or
sulphides. The occurrence of abundant swamps or marshes around or in
close proximity to a lake may restrict trace element movement into
the lake itself. By contrast, lakes in Shield areas above the
treeline and in the alpine Cordilleran regions are fed by waters
derived mainly from snowmelt and consequently contain very little
dissolved organic material. Here, adsorption of metals directly onto
clays, rock flour, and hydrous metal oxides and dissolution of
mineral particles are the predominant water-sediment interactions.
Under oxidizing conditions, hydrous oxides of iron and manganese are
excellent scavengers of trace elements; however, under reducing
conditions they are solubilized and may result in increases in
concentrations of cations and anions in overlying waters. Iron and
manganese oxides are certainly important species in organic systems
but their role as direct adsorbers/absorbers of metal ions can be
overshadowed by competition from the more reactive humic materials
and organo-clays or obscured by coatings of organic matter.
Moreover, these oxides are unstable in reducing organic-rich
sediments.
The relative amounts of organic and inorganic sediment as well as
sedimentation rates vary markedly with latitude. Stratigraphically,
lake-bottom sediment generally consists of an upper strata of
gel-like sediment, containing a variable quantity of organic
material overlying glacial lake or other glacigenic deposits. This
modern organic sediment is most widespread, thickest and
organic-rich in lakes of the southern Shield; in lakes of the
northern Shield it is deficient in organic matter, relatively thin,
areally restricted, and sometimes absent. In contrast to centre-lake
sites, material from the mineral sediment found around the margins
of lakes is often composed of glacial, glacial-lacustrine or marine
sediments, or soils which have been subjected to some reworking,
including wave action and, in those lakes in the north of the
Shield, to wind action and various periglacial processes.
Most trace metals tend to be enriched in the modern organic
sediments relative to inorganic sediments, a factor which is most
probably due to the nature of the metal-organic binding strength and
perhaps increased ion-exchange capacity of organic sediments over
inorganic types. As a result, the highest and most uniform
concentrations of trace metals generally occur in the modern-organic
sediment found in the deep central areas (profundal basins) of most
lakes (Coker et al., 1979).
The ranges of physicochemical, limnological and ecological
conditions in lakes emphasize the complexity of the lacustrine
environment and are a reflection of differences in geographic,
climatic and geological environments.
Water sampling is often an integral part of lake sediment surveys as
knowledge of the distribution of many elements in the sediments
often needs to be supplemented by information on their distribution
in the overlying waters. This additional information can often
provide some insight into the effects of variations in certain
physicochemical factors (pH, Eh, alkalinity, Mn, Fe, nature and
amount of suspended clastic and organic matter, etc.) which might
inhibit or prolong the dispersion of a given trace element in
solution in the lacustrine environment.
The presence of oxygen, carbon dioxide, methane and/or hydrogen
sulphide in their various gaseous and dissociated ionic forms,
determine whether aerobic oxidizing or anaerobic reducing conditions
exist through the water column. The nature of the electrochemical
conditions of the water column determines whether dissolved trace
metals remain in solution, or are precipitated and accumulate in the
underlying sediment, and whether trace elements are retained within
the surface sediments, or released through dissolution or desorption
back into the overlying water.
To illustrate the nature and variety of physical, chemical and
biological mechanisms that affect the distribution of trace and
minor elements, including mercury, in lake sediments a number of
examples of research carried out by GSC scientists in various areas
across Canada are discussed.
In eastern Ontario, between Ottawa/Kingston and Georgian Bay, within
the Boreal Shield ecoregion underlain by Phanerozoic sediments and
igneous and metamorphic bedrock of the Grenville Structural Province
of the Canadian Shield, concentrations of Hg and other trace
elements in lake sediment and glacial sediments can be related to
glacial dispersal from mineralized bedrock and/or bedrock with high
natural background concentrations of these elements (Coker et al.,
1995; Kettles and Shilts, 1994; and, Kettles et al., 1991). Lake
sediment cores collected and analyzed (trace and minor elements,
pollen, diatoms, Pb 210 dates, sediment composition, sedimentation
rate, local development history) from this region illustrate that
the local activities of man, and/or nature (e.g. beavers), in
clearing land and damming waterways affect the metal distribution in
sediment sequences. One doesn't necessarily see any marked
enrichment at the surface, for either Hg or other elements, but
higher levels often occur elsewhere in the sediment sequence and can
be correlated with changes in limnological environment and/or rate
of sedimentation and type of sediment deposited associated with
changes around the larger drainage basin.
The Kaminak Lake area, District of Keewatin, N.W.T., lies within the
Southern Arctic ecoregion and the Churchill Structural Subprovince
of the Canadian Shield. A detailed study of Hg in the waters of
Kaminak Lake and nearby smaller lakes showed consistent enrichment
of Hg in the waters of lakes along the trend of the sulphide-bearing
metasedimentary and metavolcanic rocks of the Hurwitz Group (Shilts
and Coker, 1995; Hornbrook and Jonasson, 1971). These natural
geologically controlled elevated levels of Hg in the lake waters
were reflected by elevated levels of Hg in fish in Kaminak Lake
(Sherbin, 1979). Within the District of Keewatin, N.W.T. several
lakes and their sediment columns were studied in some detail to
examine the geological and limnological evolution of the lake basins
since glaciation (Edwards et al., 1987; Klassen et al., 1983; and
Shilts et al., 1976). This work showed that: the rate of
sedimentation in the lakes of this area is extremely low; the lakes
are highly oxygenated with intense oxidation of the sediment at the
sediment-water interface; chemical sediments are widespread and
significant components of the sediment sequence; the organic content
of the sediment is low (<10%) and generally confined to the modern
fresh water sediments; and, there is significant variation in the
nature of the minor and trace element profiles through the lake
sediment columns related to sediment type and deposition history,
both within and between lakes. The Kaminak Lake area is remote from
anthropogenic sources and yet some of the cored sediments show that
there is an upwards increase in trace and minor elements which
commences well before industrialization.
Tatin Lake, B.C is located in the Montane Cordillera ecoregion and
Canadian Cordillera. Detailed limnological studies and lake sediment
coring of three basins within Tatin Lake (Friske, 1995) showed that
limnological variations among the lake basins are a significant
factor in controlling the nature and pattern of accumulation and
distribution of minor and trace elements at the sediment-water
interface and throughout the lake sediment columns.
Two long cores (103 from the north basin; 122 from the south basin)
collected as part of a cooperative project examining the geology and
sediment composition of Lake Winnipeg (Henderson, 1995; L. Lockhart,
pers. comm.), revealed a surface enrichment of Hg close to the
sediment-water interface and a correlation of Hg with organic matter
content within the post-glacial recent Lake Winnipeg sediment
sequence. As well, it was noted that the highest concentrations of
some minor and trace elements occurred at depth within the
underlying glacial Lake Agassiz sediments, in some instances peaking
at or close to the contact between the Lake Agassiz and Lake
Winnipeg sediments dated at around 8000 yrs. B.P. (H. Thorleifson,
pers. comm.). This suggests that factors other than those related to
anthropogenic sources play a role in controlling trace element
distribution in these cores.
Regional lake sediment and water data around the north shore of Lake
Superior, Ontario, illustrate the strong and predictable influence
of geology on the local and regional patterns of minor and trace
element distributions in the lake sediments and waters (Coker and
Shilts, 1979). Recent work in this area included the coring of: two
anomalous lakes (400 and 600 ppb Hg) within the Rove shale, a
bedrock formation demonstrated to be enriched in Hg, Zn, As, and Cd;
and, one relative "background" lake (100 ppb Hg), some 25 kilometres
away, within metavolcanic and metasedimentary bedrock having much
lower trace metal levels. The results obtained confirmed the
regional trace element levels previously obtained and showed very
consistent trace element levels from the bottom to the top of the
three cores, with no enrichment at the sediment-water interface.
This work verified the results of the original National Geochemical
Reconnaissance Survey (Geological Survey of Canada, 1978) and
confirmed the overwhelming geological control on the trace element
concentrations of the lake sediments in this area.
The examples cited illustrate the variety of physical, chemical and
biological processes, and their potential relationships and
interactions, that can effect the distribution of trace and minor
elements, including mercury, in lake waters and sediments. They also
illustrate the wide range of variation in the levels of mercury and
other elements found in lake waters and sediments across Canada.
This variation occurs within and between lakes as well as on a
regional scale and these variations can be explained in a geological
context.
There is clearly still much work to be done to understand the
processes operative on minor and trace elements within the
lacustrine environment itself, and within lakes located in different
physiographic, climatic, geographic and geological environments.
Knowledge of the processes by which a metal is mobilized,
transported, precipitated, and possibly remobilized, is of prime
concern in order to comprehend possible controls on that metal's
dispersion, accumulation and fixation into lake bottom materials.
Surface enrichment in lake sediment columns is common in
terrestrially derived minor and trace elements known to originate
from geological sources within the drainage basin. Therefore, the
phenomenon of surface enrichment does not uniquely identify
anthropogenic sources nor does it uniquely identify atmospheric
deposition as the dominant pathway. Conclusions implied from the
shape of vertical sediment profiles need to be more carefully
examined in the context of the vast array of data available in the
geoscientific and limnological literature.
Selected References
Cline, J.T. and Upchurch, S.B., 1973: Mode of heavy metal migration
in the upper strata of lake sediment: In Proceedings of the 16th
Conference on Great Lakes Research, 1973, International Association
on Great Lakes Research, p. 349-356.
Coker, W.B. and Shilts, W.W., 1979: Lacustrine geochemistry around
the north shore of Lake Superior: Implications for the evaluation of
the effects of acid precipitation: In Current Research, Part C,
Geological Survey of Canada, Paper 79-1C, p. 1-15.
Coker, W.B., Hornbrook, E.H.W. and Cameron, E.M., 1979: Lake
sediment geochemistry applied to mineral exploration: In Geophysics
and Geochemistry in the search for Metallic Ores; Peter J. Hood,
editor; Geological Survey of Canada, Economic Geology Report 31, p.
435-478.
Coker, W.B., Kettles, I.M. and Shilts, W.W., 1995: Comparison of
mercury concentrations in modern lake sediments and glacial drift in
the Canadian Shield in the region of Ottawa/Kingston to Georgian
Bay, Ontario, Canada: Water, Air and Soil Pollution, 80, p.
1025-1029.
Edwards, T.W.D., Klassen, R.A. and Shilts, W.W., 1987: Terrain
geochemistry surveys, permafrost studies, and arctic limnology,
District of Keewatin, N.W.T.: Implications for water quality
monitoring in the north: Canadian Journal of Water Pollution
Research, v. 22(4), p. 505-517.
Farmer, J.G., 1991: The perturbation of historical pollution records
in aquatic systems: Environmental Geochemistry and Health, v. 13(2),
p. 76-83.
Friske, P.W.B., 1995: Effects of limnological variation on element
distribution in lake sediments from Tatin lake, central British
Columbia - implications for the use of lake sediment data in
exploration and environmental studies: In Current Research 1995-E,
Geological survey of Canada, p. 59-67.
Geological Survey of Canada, 1978: Regional lake sediment and water
geochemical data, Ontario 1977, NTS 52A, 52H (S1/2): Geological
Survey of Canada, Open File 507.
Gorham, E. and Swaine, D., 1965: The influence of oxidizing and
reducing conditions upon the distribution of some elements in lake
sediments: Limnology and Oceanography, v. 10, p.268-279.
Henderson, P.J., 1995: The geochemistry of Lake Winnipeg long cores
and bottom sediment samples: Namao cruise 94-900: Geological Survey
of Canada, Open File Report 3113.
Hornbrook, E.H.W. and Jonasson, I.R., 1971: Mercury in permafrost
regions: Occurrence and distribution in the Kaminak lake area,
N.W.T.: geological Survey of Canada, Paper 71-43, 13 p.
Hutchinson, G.E., 1957: A Treatise on Liminology; Volume 1,
Geography, Physics and Chemistry: Wiley and Sons, New York, 1015 p.
Klassen, R.A., Matthews Jr., J.V., and Philips, L.K., 1983: Taxa in
lake sediments of the District of Keewatin: In Current Research,
Part A, Geological Survey of Canada, Paper 83-1A, p. 357-361.
Kettles, I.M., Shilts, W.W. and Coker, W.B., 1991: Surficial
geochemistry south-central Canadian Shield: Implications for
evironmental assessment: In Geochemical Exploration 1989, Part II
(A.W. Rose and P.M. Taufen, Editors), Journal of Geochemical
Exploration, v. 41 (1/2), p. 29-57.
Kettles, I.M. and Shilts, W.W., 1994: Composition of glacial
sediments in Canadian Shield terrain, southern Ontario and southwest
Quebec: application to acid rain research and mineral exploration:
Geological Survey of Canada, Bulletin 463, 58 p.
Mackereth, F.J.H., 1965: Chemical investigation of lake sediments
and their interpretation: Proceedings of the Royal Society, B161, p.
285-309.
Mortimer, C.H., 1941: The exchange of dissolved substances between
mud and water in lakes. Part I and II: Journal of Ecology, v. 29, p.
280-329.
Mortimer, C.H., 1942: The exchange of dissolved substances between
mud and water in lakes. Part III and IV: Journal of Ecology, v. 30,
p. 147-207.
Mortimer, C.H., 1971: Chemical exchanges between sediments and water
in the Great Lakes - speculations on probable regulatory mechanisms:
Limnology and Oceanography, v. 16(2), p. 387-404.
Rasmussen, P.E., 1994: Current methods of estimating mercury fluxes
in remote areas: Environmental Science and Technology, v. 28 (13),
p. 2233-2241.
Ruttner, F., 1963: Fundamentals of Limnology: University of Toronto
Press, Toronto, 295 p.
Sain, K.S. and Neufeld, R.D., 1975: A dynamic model of
biogeochemical cycle of heavy and trace metals in natural aquatic
systems: Paper presented at the Second International Symposium on
Environmental Biogeochemistry, Burlington, Ontario, Canada, 20 p.
Sherbin,I.G., 1979: Canadian Environmental Protection Service Report
EPS3-EC-79-6,359p.
Shilts, W.W., Dean, W.E. and Klassen, R.A., 1976: Physical, chemical
and stratigraphic aspects of sedimentation in lake basins of the
eastern arctic shield: In Report of Activities, part A, Geological
Survey of Canada, Paper 76-1A, p. 245-254.
Shilts, W.W. and Coker, W.B., 1995: Mercury anomalies in lake water
and in commercially harvested fish, Kaminak Lake area, District of
Keewatin, Canada: Water, Air and Soil Pollution, 80, p. 881-884.
Walters, L.J., Herdendorf, C.E., Charlesworth, J.L., Anders, H.K.,
Jackson, W.B., Skoch,E.S., Webb, D.K., Kovacik, T.L. and Sikes,
C.S., 1972: Mercury contamination and its relation to other
physiochemical parameters in the western basin of Lake Erie: In
Proceedings of the 15th Conference on Great Lakes Research, 1972,
International Association on Great Lakes Research, p. 306-316.
[INLINE]
back to intro next section
BACK TO
*********************************************************************