Journal of Minerals & Materials Characterization & Engineering, Vol. 2, No.2, pp 111-135, 2003
http://www.jmmce.org, pri nte d i n t he USA . A ll rig hts re ser ved
111
Characterization and Transport of Contaminated Sediments in the Southern Central Lake
Superior
Jaebong Jeong1* and S. Douglas McDowell2
1Institute of Materials Processing
2 Geological and Mining Engineering & Sciences
1400 Townsend Dr. Houghton, MI 49931
*Corresponding author. Tel.: -906-487-2887
Fax: -906-487-2943
E-mail address: jjeong@mtu.edu
Three major source sediments were characterized and classified in terms
of mineralogical and chemical composition in the west coastal area of the
Keweenaw Peninsula. Bulk chemical analysis reveals that concentrations of Cu,
Ag, Co, and As were enriched in metal rich mine tailings. SEM-EDS analysis
indicates that the Ontonagon River sediments have high P and S concentrations.
X-ray diffraction analysis of clay fraction shows that the mine tailings (chlorite
rich) could be distinguished from the other two sources, Ontonagon River
sediments (low chlorite and high illite) and Wisconsin red clay (low illite and
high expandable phase). Local environmental conditions, including currents,
bathymetry, weather conditions, and sediments texture, are important factors for
cross-margin and longshore transport of contaminated sediments. The Keweenaw
Current is responsible for the longshore transport of fine fraction of tailings,
whereas wave action causes the lateral transport of the coarse deposits along the
shore.
Key words: tracer, metal fingerprint, particle transport, and trace metal.
1. Introduction
Particulate matter is well known to be a source, sink, and a transport vehicle of nutrients
and contaminants in lake systems (Bahnick et al., 1978; Stumm and Morgan, 1996).
Consequently particulate matter plays an important role in controlling the chemical composition
of surface waters and sediments via a complex combination of biological, chemical and physical
processes (Honeyman and Santsch, 1989; Lion et al., 1982; Martin et al., 1995; Morel and
Hudson, 1985; Ribolzi et al., 2002; Sigg, 1985; Sunda and Huntsman, 1995; Whitfield and
Turner, 1987). Different types of particulate matter originating from different areas and
processes are helpful in assessing the sediment budgets of lake systems (Colman and Foster,
1994; Klump et al., 1989). Therefore, study of the origin and movement of particles is useful to
assessing sediment loads and provenance in depositional zones, and understanding issues such as
nutrient levels, primary productivity, contaminant movement, and allowable discharge limits in
lake systems.
112 Jaebon g Jeong and S . Do u g las M c D owell Vol.2, No.2
A previous study of sedimentation in Lake Superior found that the major sources of
sediments in depositional zones are coastal erosion and tributary inputs (Kemp et al., 1978a).
However, this study found a discrepancy in the sediment budget for the lake; the sparse
measurements of recent sedimentation rates were much lower than the estimates of sediment
inputs. The study also was unable to link specific sediment sources with the sediments deposited
in each depositional basin. An understanding of the transport behavior of the source materials
within the lake could explain the discrepancy in the sediment budget for the lake. Sedimentation
in each depositional basin is strongly related to the combined effects of localized inputs and
physical processes (e.g., currents, wave action). The eastern basin sediments containing a
relatively high content of SiO2 are derived from Ontario soils containing more illite, whereas the
extreme western sediments derived from Manitoba, Canada contain more expandable clays
(Dell, 1973; Forman and Brydon, 1961; Nussmann, 1965; Thomas and Dell, 1978). In the
Keweenaw Peninsula region of Lake Superior, Wisconsin red clays from shoreline erosion
(Kemp et al., 1978a), sediment loads from the Ontonagon and Bad Rivers (Auer and Gatzke,
2002; Kemp et al., 1978a; Robertson, 1997), and copper mine tailings (Babcock and Spiroff,
1970) are the major sources of sediments. Although these sources have been identified and
quantified, the transport behavior and the fate of these materials in the lake are still unknown
(Churchill et al., 2002).
Several different techniques have been applied for particle tracking to understand the
biogeochemical cycling of nutrients, the fate of pollutants associated with particles, and
sedimentation budgets in limnetic systems. In Lake Superior, hydrophobic organic contaminants
such as PCBs and PAHs have been used as tracers of the dynamics and transport of organic
particles in large lakes (Baker and Eisenreich, 1989). The elemental and isotopic compositions of
suspended particles and sedimentary organic matter also have been used to study the origin and
cycling of these materials (Ostrom et al., 1998). Even rare earth elements were applied as tracers
to study sediment reworking and transport in the depositional basin (Krezoski, 1989). In
addition, mineralogical and chemical compositions of particles and sediments have been used to
identify sediment sources and to understand the impact of human activities on local
environments. X-ray diffraction and Transmission Electron Microscopic studies of suspended
particles and lake sediments identified mineralogical composition and particular minerals
(asbestiform amphibole fibers) that could be used as fingerprints to classify sediment sources and
track taconite tailings (Cook, 1975; Cook and Rubin, 1976; Dell, 1973). In the Keweenaw
Peninsula region of Lake Superior, elevated concentrations of trace elements and high Cu/Zn
ratios in sediments have been used as tools to determine the perturbation of local ecosystems by
the tailings discharged from copper mining activities (Kemp et al., 1978b; Kerfoot et al., 1994;
Kerfoot and Robbins, 1999; Kolak et al., 1998).
The objectives of this study were to examine transport and redistribution of coastal
sediments in the Keweenaw Peninsula. Specifically, “fingerprints” of the different sediment
source materials were developed using mineralogical and chemical compositions (major element
and trace metal contents) in order to distinguish the sources from one another. Longshore and
cross-margin transports of particles and redistribution of sediments were then investigated to
understand the fate of coastal and riverine inputs in the study area. Finally, factors controlling
longshore and cross-margin transport of particles and redistribution of sediments were identified
with respect to local factors such as wind-driven coastal currents, waves, and bathymetry.
Vol.2, No.2 Charact erizatio n and Tr ansport of Conta minated Sediment s in the Lake Superior 113
113
2. Methodology
2.1 Site description
The study site is located along the west coast of the Keweenaw Peninsula in the southern
central basin of Lake Superior. Bathymetry in the area varies from a shallower western shoreline
to a deeper northern shoreline, which has the effect of intensifying the water circulation pattern
(Van Luven et al., 1999). The major source of fine-grained particles in the study area is the red
clay from the Wisconsin shoreline (Figure 1). The second largest source is riverine inputs (Kemp
et al., 1978a; Robertson, 1997). In addition to shoreline erosion and river loads, copper mine
tailings from Freda, MI are also one of the major sources of sediments in the study area due to
the discharge of over 45 million metric tons of crushed rocks directly into the lake (Babcock and
Spiroff, 1970). The study area extends from the Ontonagon River to Copper Harbor, Michigan,
the Keweenaw Peninsula region of Lake Superior. Five sampling transects were located at
Ontonagon (ON Transect), near Freda and Redridge (FR), at the north entry of the Portage
waterway (HN), at Eagle Harbor (EH), and at Copper Harbor (CH) (Figure 1). Transects run
perpendicular to shore in a northward direction (305o ~ 350o), and each has several sampling
stations.
2.2 Sample collection and preparation
Samples of three source materials, forty-two lake sediment, one sediment core, and four
suspended sediment traps were taken over a three-year period from 1998 to 2000. The three
source sediments are (1) Wisconsin red clays from the shoreline bluffs at the mouth of the Poplar
River in northern Wisconsin, (2) Ontonagon River sediments from the mouth of Ontonagon
River in Michigan, and (3) stamp sands from Freda-Redridge in Michigan. Forty-two lake
sediment samples were taken along the study transects at distance intervals of 0.5 to 5 km. One
sediment core was taken near Copper Harbor (MCA2) using a Multicorer. The first 10 cm of the
core were sliced at 0.5-cm increments and the second 20 cm were sliced at 1-cm increments on
the boat. Three samples were prepared for analysis: surface sediment (0.5-1 cm), sediment (2.5-3
cm) from maximum Cu concentration, and background sediment (9-9.5 cm). For collection of
settling particles, four trap moorings were deployed at water depths of 50 and 120 m on the HN
transect. Samples were stored in polyethylene bags at 4oC until analyzed.
114 Jaebon g Jeong and S . Do u g las M c D owell Vol.2, No.2
Ontario
Ontario
Minnesota
Wisconsin
Michigan
Lake Superior
ite (MCA2)
Ontonagon Sediment
WI Red Clay
Freda Stamp sands
46.75
47.00
47.25
47.50
47.75
87.588.088.589.089.5
Longitude
Latitude
ON Transect
HN Transect
Ontonagon
Freda
FR TransectCopper
Harbor
Eagle
Harbor
Redridge
Keweenaw Peninsula
EH TransectCH Transect
46.75
47.00
47.25
47.50
47.75
87.588.088.589.089.5
Longitude
Latitude
ON Transect
HN Transect
Ontonagon
Freda
FR TransectCopper
Harbor
Eagle
Harbor
Redridge
Keweenaw Peninsula
EH TransectCH Transect
ON Transect
HN Transect
Ontonagon
Freda
FR TransectCopper
Harbor
Eagle
Harbor
Redridge
Keweenaw Peninsula
EH TransectCH Transect
Figure 1
Wet sediment samples were divided into three portions that required different preparatory
treatments. The first fraction was dried and used for mineralogical analysis using a Siemens
D500 X-ray diffractometer (XRD). For major mineral identification, approximately five grams
of each sample were pulverized using an industrial-grade blender with iron beads. For clay
mineral analysis (<2 µm), the clay fraction was separated by gravity sedimentation, concentrated
with a centrifuge, and evenly coated on the surface of a glass microscope slide. Slide samples
were dried under a clean air environment for 24 hours, x-rayed, and exposed to ethylene glycol
vapor in a closed ethylene glycol desiccator and re x-rayed. The second fraction was completely
dried at 105 oC and was used for individual particle analysis using a Scanning Electron
Microscope (SEM, JSM-35C, JEOL) and an Electron Microprobe Analyzer (EMPA, JXA-8600,
JEOL). For bulk chemical analysis, the third fraction was dried at 105 oC, whereas sediment trap
samples were dried in a benchtop freeze dry system (FreeZone 6 Liter Benchtop, Model 77520,
Labconco). The sediment core sample was processed at the Large Lakes Observatory (LLO) of
the University of Minnesota-Duluth. The third fraction and core samples then were digested
using a one-step extraction (OSE) method to measure total concentrations of trace elements in
the samples (Jeong et al., 1999).
Vol.2, No.2 Charact erizatio n and Tr ansport of Conta minated Sediment s in the Lake Superior 115
115
2.3 Particle size, chemical, and mineralogical analyses
The grain size distribution of sediments was obtained using sieves and an X-ray particle
analyzer. Four U.S Standard Sieves, 12-, 20-, 40-, 70-, and 100 mesh (1680-, 840-, 420-, 210-,
and 149 µm respectively) were used with dry sediments to obtain coarse particle size
distributions. The fine fraction (less than 149 µm) was analyzed with an X-ray particle analyzer
(MICROTRACK II Model 7997-10, Leeds & Northrup). Each sample was analyzed twice, and
the means were accurate within 5% of the particle size.
For mineralogical analysis, both air-dried and ethylene-glycol-treated clay-size particles
were examined over a range of 2θ from 2
o to 65o at a scanning rate of 1.00 Deg/min using a Cu
Kα radiation. For individual particle analysis, the carbon-coated specimens were analyzed using
an automated computer program (FeaturescanTM, Link Analytical) on a JOEL 35C SEM and a
JOEL 8600 EMPA interfaced with an energy-dispersive spectrometer (EDS) system.
Approximately 300 particles per sample were examined using both SEM and EMPA. Chemical
microanalyses of individual particles were performed for the following elements: Na, Mg, Al, Si,
S, K, Ca, Ti, Fe, and Cu. Using the chemical composition data for individual particles obtained
from SEM and EMPA, two multivariate statistical analyses (linear discriminant analysis and
logistic regression) were performed for the three source materials in order to identify the unique
characteristics of each source. Statistical analyses were applied to both the large particles (<149
µm) analyzed with SEM and small particles (<2 µm) analyzed with EMPA using morphological
and chemical variables.
Major elements such as Fe, Mg, Ca, Mn, Na, Si, and K in extracts from OSE were
analyzed using an Inductively Coupled Plasma Emission Spectrometer (ICP, Leeman Labs Inc.).
Trace elements (V, Co, Ni, Zn, As, Sr, Ag, Cd, Rb, Ba, Pb, and U) were analyzed using an
Inductively Coupled Plasma Mass Spectrometer (ICP-MS, Perkin-Elmer Elan-6000, in Duluth,
MN). Concentrations of total copper were measured using an Atomic Absorption
Spectrophotometer (Perkin-Elmer Co., Model AAS 3100) in flame mode (3100 Automatic
Burner Control). A NIST standard reference material (SRM 2704: Buffalo River Sediment) was
digested and analyzed; the range of recovery was between 76% (Pb) and 89% (Co) total metal.
The reproducibility of triplicate measurements for major and trace elements was generally better
than 30%. For total copper analysis, OSE was applied to the NIST standard reference material
(SRM 2704: Buffalo River Sediment) and yielded a recovery within 10% of the certified
concentration of copper. The accuracy of copper standards was checked against a NIST standard
reference material (SRM 3172a Multielement Mix B-1); standards were within 6 % of the
certified standard. The procedural blank generally had a value lower than the detection limit.
3. Results
3.1 Sediment Grain Size
The sediment source materials and lake surface sediments have a wide range of particle
sizes (Table 1). Comparison of the fine fraction of the three source sediments indicates that Freda
stamp sands are predominantly sand and silt, Ontonagon River sediments are mostly silt, and
Wisconsin red clays in shoreline bluffs have silt and clay sized particles. The predominant
particle size of stamp sands in the Keweenaw Peninsula region is coarse sand (Kennedy, 1970).
The surficial sediments in the Nemadji River basin have more clay-size particles than those in
the Ontonagon River basin (Robertson, 1997). Surface sediments in nearshore areas are
relatively coarse compared to those in offshore areas.
116 Jaebon g Jeong and S . Do u g las M c D owell Vol.2, No.2
Table 1. Particle size distribution of three source sediments from tributary, and surface sediments and
suspended particles from Lake Superior.
Tributary Lake Superior
Particle
Size Sources Surface Sediment Suspended
Particles6
Size\Site Freda
Stamp
Sands1
Ontonagon
River
Sediments2
WI Red
Clay3 Nearshore4 Offshor
e5 All
% Clay (<2 µm) 0.0 3.0 22.5 0.0 - 1.6 0.0 -
9.7
% Silt (2<X<60
µm) 46.5 89.2 77.5 0.0 -7 1.1 18.6 -
97.4
% Sand (> 60
µm) 53.5 7.8 0.0 27.4 -
100.0 0.0 -
81.4
Mean ( µm)7 66.9 24.1 5.6 460.0 61.0 4.2-5.0
1) From Freda, MI and less then 200 mesh size particle (U.S. standard sieve).
2) From the river mouth, and less then 200 mesh size pa rticle (U.S. standard sieve).
3) From the shorel ine bluffs at the mouth of the Poplar R iver in north ern Wisconsin .
4) Depth is shallower then 60 m.
5) Depth is deeper then 60 m.
6) Measured usi ng Scannin g Electron Microscopy.
7) Mas s mea n diam e t e r
3.2 Mineralogical and chemical analyses
X-ray diffraction analyses of the source materials and Lake Superior sediments showed
quartz, illite, chlorite, and smectite to be the dominant mineral components. Minor calcite is
present in the Freda stamp sands and Wisconsin red clay, but not found in the Ontonagon River
sediments. Lake Superior sediments generally do not contain calcite due to the undersaturation of
carbonates in the lake water (Thompson, 1978). Smectite (montmorillonite), feldspar (microcline
and albite), mica (muscovite), kaolinite, epidote, titanite, and augite, were found as minor
minerals in some of the sediments. Several oxidized copper minerals (tenorite, malachite, and
chalcopyrite) were found in the Freda stamp sands along with native copper.
Analyses of ethylene-glycol-treated samples of the fine material in the source sediments
showed that expandable mixed-layered clay minerals such as smectite/illite are well separated
from the crystalline, or non-expandable, chlorite peak at low 2θ angles in the Wisconsin red clay
and Ontonagon clay (Figure 2a and b). Freda stamp sands are dominated by chlorite with minor
illite and trace amounts of expandable clay, whereas Wisconsin red clays contain significant
amounts of expandable clay, illite, and chlorite. The clay fraction of the Ontonagon River
sediments mainly contained expandable clay minerals (smectite and illite) and were similar to
the Wisconsin red clay but have lesser amounts of chlorite. The clay-size lake sediments also
contained expandable clay minerals and crystalline chlorite. Source and lake sediments were
characterized with respect to clay minerals using the ratios of the major peak intensities after
background subtraction (Table 2). These ratios then were used in a classification scheme for the
clay-size sediments retrieved from Lake Superior.
Vol.2, No.2 Charact erizatio n and Tr ansport of Conta minated Sediment s in the Lake Superior 117
117
0
200
400
600
800
1000
1200
1400
26 10 14 18 22 26 30
2 Theta
Intensity
Freda stamp sands
Ontonagon clay
Wisconsin red clay
Chlorite 1
Chlorite 2
Chlorite 3
Chlorite 4
Smectite
illite 1
illite 2
illite 3
Albite
Figure 2a
0
200
400
600
800
1000
1200
1400
26 10 14 18 22 26 30
2 Theta
Intensity
Figure 2b
118 Jaebon g Jeong and S . Do u g las M c D owell Vol.2, No.2
Table 2. Ratios of the relevant peaks for the source materials and lake sediments.
Peak Ratio 7
Sediment Classification
ChloriteSmectite
Smectite
II
I
+
IlliteSmectite
IlliteII
I
+
ChloriteIllite
Chlorite
II
I
+
Fr eda St amp Sands 0.08 0.72 0.82
Ont on a g o n C l a y 0.30 0.60 0.61
Source Sediment
WI Red Clay 0.44 0.44 0.62
FR 0011 0.07 0.74 0.81
FR 0202 0.25 0.73 0.52
Near Freda Sediment
FR 0703 0.46 0.44 0.60
MCA-Pre4 0.20 0.75 0.57
MCA-Cu5 0.18 0.72 0.64
Near Copper Harbor
Core Sample
MCA-BG6 0.49 0.45 0.56
1. Freda 001: 0.1 km away from shore in Freda Transect.
2. Freda 020: 2 km away from shore in Freda Transect.
3. Freda 007: 7 km awa y from shore in Freda Transect.
4. Surface sediment (0.5-1 cm) from the Multiple core station 2 (MCA in Figure 1).
5. Sediment (2.5-3 cm) from maximum Cu concentration in the Multiple core.
6. Background sediment (9-9.5 cm) in the Multiple core.
7. 2θ values for smectite, chlorite, and illite are 5.40, 6.35, and 9.00 respectively.
Sediments from the shore of Lake superior contain similar concentrations of the major
elements because, with the exception of the mine tailings (Table 3), they have similar origins in
the regionally distributed reddish brown or grayish postglacial surface sediments (Babcock and
Spiroff, 1970; Laberge, 1994). Among the source materials, the Wisconsin red clays are
distinctly lower in Na. The Freda stamp sands are generally similar to the other sediment
sources, so that the major elements are not useful in tracking sediment movement in the lake. In
general, all lake sediments are distinctly higher in V and Ni, and slightly higher in Pb, then any
of the source materials (Table 4). This may reflect the highly refractory nature of the mineral
phases containing the immobile trace elements, which tend to concentrate in the sediments
during weathering. All sediments are similar in their Co, Zn, and As contents, including the
Ontonagon and Wisconsin red clay sources, while the Freda stamp sands have distinctly higher
concentrations of Co, Cu, As, and Ag.
Copper concentrations in the sediment source materials and lake sediments have a clear
pattern (Table 4). High total copper concentration was found in the Freda stamp sands (5270
µg/g), and low concentrations were found in the Ontonagon River sediments (60 µg/g) and
Wisconsin red clay (70 µg/g). In the surface sediments off Freda, copper concentrations in the
nearshore sediments (430 ± 180 µg/g) were higher than in offshore sediments (220 ± 70 µg/g).
Relatively lower copper concentrations (170 µg/g) were found on the ON transect, and relatively
high concentrations (500 µg/g) were found in the Houghton transect sediments. The modern
sediment trap samples obtained on the HN transect had a relatively high copper concentration
(160 µg/g) compared to the sediment source materials from non-mining areas. The deep, pre-
Vol.2, No.2 Charact erizatio n and Tr ansport of Conta minated Sediment s in the Lake Superior 119
119
mining sediments of the core (MCA-BG) collected near Copper Harbor had the lowest
concentration of copper (30 µg/g) found in the lake, but the sediments at the surface of this core
had a relatively high concentration (110 µg/g). The maximum copper concentration (180 µg/g) in
the core was found at 2 2.5 cm depth sediments, which are around 100 year old; this is thought
to correspond to the maximum discharge of mine tailings in the Freda and Redridge areas during
the copper mining activity from 1860 to 1960 (Babcock and Spiroff, 1970; Kerfoot et al., 1994).
Overall, elevated copper concentrations were found at the surface and at 2.5 cm depth in the core
as well as in settling particles on the HN transect, whereas the two non-mining sediment sources
had copper concentrations similar to sediments below 5 cm in the core.
Table 3. Concen trations ( mg/g) of majo r elements in sediments (Mean and standard dev iation in parentheses).
Type Sample Site Fe Mg Ca Mn Na Al Si K
Freda Stamp
Sand 45.8
(6.7) 0.02
(0.02) 0.25
(0.07) 0.65
(0.17) 5.3 (2.6) 15.6
(8.7) 228.2
(45.3) 11.2
(4.1)
Ontonagon
Sediment 21.6 0.01 0.19 0.43 6.4 22.6 236.3 14.8
Source
WI Red Clay1 41.3 0.13 0.4 0.53 0.6 19.5 205.6 19.8
Settling
Particle HN Transect2 35.5
(11.8) 0.09
(1.05) 0.26
(0.76) 0.53
(0.10) 4.1 (2.0) 21.6
(6.2) 242.0
(38.7) 15.7
(2.0)
MCA-Surf3 22.8 0.09 0.35 0.37 6.3 14.8 277.5 10.5
MCA-Cu4 18.2 0.10 0.36 0.26 5.0 11.5 209.7 8.6
Core
Sediment
MCA-BG5 16.7 0.09 0.36 0.32 5.5 14.3 319.1 11.3
HN Offshore 33.3
(26.0) 0.02
(0.00) 0.21
(0.09) 0.52
(0.34) 4.3 (0.9) 28.8
(14.5) 301.6
(73.7) 17.2
(1.32)
FR & RR
Nearshore6 46.4
(11.2) 0.12
(0.04) 0.62
(0.20) 0.63
(0.19) 5.1 (1.5) 10.4
(2.0) 185.7
(45.6) 9.7
(3.4)
Surface
Sediment
FR & RR
Offshore 45.4
(15.1) 0.15
(0.10) 0.52
(0.21) 0.64
(0.16) 8.3 (4.1) 20.8
(11.4) 240.2
(72.5) 15.9
(7.2)
1. WI: Wisco nsin.
2. HN: Hou g h ton N orth .
3. MCA-Surf: Surface potion (0.5-1 cm) of multicore sediment.
4. MCA-Cu: Maximum Cu concentration portion (2.5-3 cm) of multicore sediment.
5. MCA-BG: Background portion (9-9.5 cm) of multicore sediment.
FR & RR: Freda and Redridge.
120 Jaebon g Jeong and S . Do u g las M c D owell Vol.2, No.2
Table 4. Concentrations (µg/g) of trace elements in different sediments (Bolds represent
extremely higher values among the three source materials).
Type Sample Site V Co Ni Cu
3 Zn As S
r Ag Cd R
b Ba Pb U
Freda Stamp
Sands 34 18
4 50 527
0 70 12.
1 3
2 1.0
4 0.2
3 0.
6 42 0.1
9 0.
1
Ontonagon
Sediment 19 66 30 60 40 1.3 1
2 0.0
5 0.1
3 2.
8 56 0.1
2 0.
4
Source1
WI Red Clay 10
8 70 39 70 38 0.5 5
0 0.0
6 0.2
5 3.
8 17
0 0.1
2 1.
5
Settling
Particle HN Transect 160 0.3
5
MCA-Surf 24
3 78 10
2 110 70 0.4 0.4
4
MCA-Cu 16
0 72 10
1 180 54 0.3 0.3
7
Core
Sediment2
MCA-BG 22
1 77 96 30 54 0.4 0.4
9
HN Transect 500 0.1
6
FR & RR
Nearshore 29
7 73 94 430 70 0.3 0.2
5
FR & RR
Offshore 31
5 65 79 220 10
0 0.3 0.2
8
Surface
Sediment2
ON Transect 170
1) Co nce ntr ati ons of trace me tal s e xce pt Cu in sou rce s w ere me asu red by IC PMS in Du lut h.
2) Co ncentrat ion s o f t rac e metals ex cep t C u i n core a nd sur fac e s ediments we re mea sur ed by IC P a t M TU.
3) Co ncen trations of Cu were m ea s ur ed b y A AS a t M TU , a nd t ho s e ar e m ea n va l ue s.
Vol.2, No.2 Charact erizatio n and Tr ansport of Conta minated Sediment s in the Lake Superior 121
121
3.3 Individual Particle Analysis
Distributions of the elements in the large and small particles from the three sediment
sources are shown in Figure 3a and Figure 3b. The box plots show that most element
concentrations are similar to one another in the large-particle source materials except for two
elements (P and S). On this basis, Ontonagon River particles could be distinguished from the
others because of their high P concentration, low S concentration, and higher ratio of Mg/Al.
Unlike the large particles, there are no significant differences in morphological or element
concentrations in the small particles from the three sources. The finer particles are noticeably
higher in Al, S, and slightly higher in Si, K, and Cu, then the coarser particles. This may reflect
in part a higher concentration of layer silicate minerals in the finer fraction.
Freda Onto WI Red
0
10
20
30
40
50
Freda Onto WI Red
0
50
100
150
200
Freda Onto WI Red
0
100
200
300
400
500
Freda Onto WI Red
0
1000
2000
3000
4000
Freda Onto WI Red
0
1000
2000
3000
4000
5000
Freda Onto WI Red
0
50
100
150
200
Freda Onto WI Red
0
100
200
300
400
500
600
Freda Onto WI Red
0
500
1000
1500
2000
Freda Onto WI Red
0
100
200
300
400
500
600
700
800
900
1000
Freda Onto WI Red
0
1000
2000
3000
Freda Onto WI Red
0
100
200
300
Freda Onto WI Red
0
1
2
3
4
5
6
7
8
Na
Mg/AlCuFeTi
CaKSP
SiAlMg
Counts
Figure 3a
122 Jaebon g Jeong and S . Do u g las M c D owell Vol.2, No.2
Freda OntoWi Red
0
10
20
30
40
Freda Onto Wi Red
0
2
4
6
8
10
12
Freda Onto Wi Red
0
10
20
30
40
50
Freda Onto Wi Red
0
50
100
150
Freda OntoWi Red
0
100
200
300
400
500
600
Freda Onto Wi Red
0
1000
2000
3000
4000
5000
6000
7000
Freda Onto Wi Red
200
300
400
500
600
700
Freda Onto Wi Red
0
1000
2000
3000
Freda OntoWi Red
0
1000
2000
3000
4000
Freda Onto Wi Red
0
1000
2000
3000
Freda Onto Wi Red
0
1000
2000
3000
4000
Freda Onto Wi Red
0
100
200
300
400
500
MgNa
K
Si S
LengthArea
Ca Ti Fe Cu
Al
Counts
Figure 3b
The two multivariate statistical analyses (linear discriminant analysis and logistic
regression) of the chemical composition data obtained from SEM and EMPA provided similar
results for particle classification in both small- and large-particle portions. For identification of
small particles, more than half of those in each source group are misclassified in both statistical
analyses. However, when applied to the large particle measurements, both statistical techniques
classified the particles correctly 60 to 70% of the Freda stamp sands and the Wisconsin red clay,
and more than 99% of the Ontonagon River sediments. The relative importance of variables was
also determined by the F-to-remove (F) statistics in the discriminant analysis and the odds ratio
in the logistic regression model. For the statistical analyses of the large particles, the order of
group differences is S > P >> Cu >>> K > Al > Mg > Na in the discriminant analysis, whereas
the orders of significance of individual parameters are Cu > Na > S > Mg> Ca > Al = Si > K for
the Freda group against Wisconsin red clay group and Na > Mg> K > Ca > Si > Cu > S > Al for
the Freda group against Ontonagon sediment group in the logistic regression model.
4. Discussion
4.1 Characterization of source particles
Many studies have attempted particle tracking using mineralogical fingerprints (Chamley,
1989; Cook, 1975; Dell, 1973; Gutierrez et al., 1996). Amphiboles in taconite tailings have been
used as an indicator of seasonal variations of water quality (Cook, 1974). Calcite and dolomite
have been used for identification of late- and post-glacial sediments, whereas compositions of
mixed-layer clays and quartz in the sediments were used to identify different riverine sources
(Dell, 1973). Apatite, the most abundant and insoluble phosphate mineral, was used for assessing
Vol.2, No.2 Charact erizatio n and Tr ansport of Conta minated Sediment s in the Lake Superior 123
123
the phosphorus loading in lake sediments from glacial debris or metamorphic terrains (Reid et
al., 1980). Mixing of riverine with marine sediments has been explained by different
mineralogical composition of clay minerals (Chamley, 1989). In a similar fashion, mineralogical
fingerprints of mine tailings were employed as tools to explore the dynamics and transport of
particle and sediments associated with local environments in this study.
The geologic history of the southern coast of Lake Superior resulted in mineralogically
similar sources of lacustrine sediments to the lake (Heinrich, 1976; Laberge, 1994). The red
clays from northern Wisconsin bluffs and the Ontonagon River sediments are glacial lacustrine
sediment derived both from older clastic sediments within the Lake superior trough, and from the
weathering of nearby basaltic volcanics. The mine tailings discharged in Freda, as a result of
copper mining activities, are amygdaloidal basalts (Babcock and Spiroff, 1970). X-ray
diffraction analyses for the major mineral phases reveals similar minerals but different mineral
proportions among the source materials and lake sediments. The copper minerals (tenorite,
malachite, and chalcopyrite) identified by X-ray diffraction analysis could not be used for
differentiation of tailings from non-mining sediments due to trace amounts compared to major
minerals. However, the clay mineralogy of the fine particle fraction indicated that the mineral
composition of each sediment source was unique and could be used to discriminate them from
one another (Figure 4). The relative abundances of illite, smectite, and chlorite for the three
different sources and the lake sediments can be used to characterize the sources and classify the
lake sediments. The stamp sands are clearly distinguished from the others by relatively low
smectite and high chlorite levels, whereas different proportions of illite and smectite
distinguished the Ontonagon River sediments from the Wisconsin red clays.
Figure 4
124 Jaebon g Jeong and S . Do u g las M c D owell Vol.2, No.2
Bulk chemical compositions (Table 3) of the three source materials differ only slightly
for some major elements. However, ratios of refractory and mobile elements are a more powerful
tool for characterizing the sources. Ontonagon River sediments are more clearly discriminated
from the others with respect to ratios of refractory (Al and Si) and mobile elements (K, Mn, Ca,
and Mg) due to higher aluminum and lower calcium and magnesium (Figure 5a). On the other
hand, high ratios of cations:Si and K:Na in Wisconsin red clay discriminated this material from
other source materials. Major elements and their ratios can be used to develop strong fingerprints
for the three source materials having different origins and undergoing different weathering
processes.
Frequently, high concentrations of trace metals in sediments are a result of mining
activity (Dassenakis et al., 1995). In this study, large differences in the abundance of trace metals
were found in the three source sediments (Figure 5b); total concentrations of trace metals such as
Cu, Ag, and As in Freda stamp sands were from 10 to 100 times higher than concentrations in
the other source materials. This was consistent with previous research that had shown that mine
tailings in the study area were clearly different from normal lake sediments with respect to high
levels of trace metals (Jeong et al., 1999; Kemp et al., 1978b; Kerfoot et al., 1994).
1.0E-01
1.0E+00
1.0E+01
1.0E+02
1.0E+03
1.0E+04
(Al+Si)/(Fe+Mn) (Al+Si+K)/(Ca+Mg) (Ca+Mg+Na)/Si K/Na
Concentration (mg/g)
Freda Stamp Sands
Ontonagon Clay
WI Red Clay
Figure 5a
1.E-02
1.E-01
1.E+00
1.E+01
1.E+02
1.E+03
1.E+04
Cu CoZn Pb As VNi Sr Rb BaUGaCd Ag
Log [Concentration] (ug/g)
Freda Stamp Sands
Onto Sediment
WI Red Clay
Figure 5b
Vol.2, No.2 Charact erizatio n and Tr ansport of Conta minated Sediment s in the Lake Superior 125
125
For individual particle analyses, the multivariate statistical analyses of major element
concentrations in fine particles of the source materials failed to discriminate one source from
another due to the similarity of the chemical composition of the source materials. However,
analysis of large particles (<200 µm) by SEM-EDS coupled with multivariate statistical analysis
succeeded in identifying Ontonagon River particles in almost 100% of cases (Figure 6). This
result was consistent with the bulk chemical analysis for major chemical elements that showed
that the Ontonagon River sediments could be identified by a high ratio of (Al+Si):(Fe+Mn),
while the other two source materials were similar (Figure 5a). The results imply that EMPA-EDS
was not an appropriate technique to obtain valuable chemical composition data to classify the
fine fraction of particles. Nevertheless, this technique could be used as an auxiliary tool to verify
the particle classifications obtained from bulk chemical analysis.
-6-226 10
FACTOR(1)
-6
-2
2
6
10
FACTOR(2)
WI Red Clay
Onto Sediment
Freda SS
-6-226 10
FACTOR(1)
-6
-2
2
6
10
FACTOR(2)
WI Red Clay
Onto Sediment
Freda SS
-6-226 10
FACTOR(1)
-6
-2
2
6
10
FACTOR(2)
WI Red Clay
Onto Sediment
Freda SS
-6-226 10
FACTOR(1)
-6
-2
2
6
10
FACTOR(2)
-6-226 10
FACTOR(1)
-6
-2
2
6
10
FACTOR(2)
WI Red Clay
Onto Sediment
Freda SS
Figure 6
Each single mineralogical or chemical analysis of the three-source sediment materials did
not provide sufficient information for classification of the sediments. However, integration of all
information from the mineralogical, major and trace elemental, and individual particle analyses
for the sediments yielded a clear fingerprint for each source material (Table 5). The next
challenge was to use these fingerprints for tracking the movement of particles in the lake.
126 Jaebon g Jeong and S . Do u g las M c D owell Vol.2, No.2
Table 5. Comparison of characteristic of the three different source materials in the study area.
Source
Analytical
Method Target
Particle Informati
on Freda
Stamp
Sands
Ontonagon
River Sediments Wisconsin
Red Clays
U.S.
Standard
Sieve1 &
MICROTRA
CK2
Whole
sample
Texture
of
sediment
Sand and
Silt Silt Silt and
Clay
Large
particles7
Major
mineral
compositi
on
Calcite with
copper
minerals No Calcite Calcite
X-ray
diffraction
Small
particles8
Clay
mineral
compositi
on
More
chlorite
with less
smectite
More illite with
less chlorite
More
smectite
with less
illite
OSE3/AAS Large
particles Total Cu High Low Low
OSE3/ICP &
ICP-MS Large
particles
Major
and trace
elements
High trace
elements
(Pb, Co,
Ag, As)
High Al and Si
with low Ca,
Mg, and low
trace elements
(Pb, Co, Ag, As)
Low Na
with low
trace
elements
(Pb, Co,
Ag, As)
EMPA/EDS4
with
Statistical
Analysis
Small
particles
Chemical
compositi
on No trend No trend No trend
SEM-EDS5
with
Statistical
Analysis
Large
particles
Chemical
compositi
on
High S and
Al with low
P, Na, and
Mg
High P, Na, and
Mg with low S
and Al
High S and
Al with low
P, Na, and
Mg
1. U.S. standard sieve: Four sieves with 12, 20, 40, 70, and 100 mesh
2. MICROTRACK: an X-ray particle size analyzer
3. OSE/AAS: One step extraction with Atomic Absorption Spectrophotometer
4. EMPA-EDS: Electron Microprobe Analyzer with an Energy-Dispersive Spectrometer
5. SEM-EDS: Scanning Electron Microscopy with an Energy-Dispersive Spectrometer
6. Radioisotope: U-235, U-248, and K-40 series
7. Large particle: Less than 200 µm size particles
8. Small particle: Less than 2 µm size particles
Vol.2, No.2 Charact erizatio n and Tr ansport of Conta minated Sediment s in the Lake Superior 127
127
4.2. Movement of copper-rich sediments
Mineralogical and chemical fingerprints of three source materials were applied to
examine transport of copper-rich sediments in the study area. The copper distribution map
clearly shows that high concentrations of copper exist between Freda-Redridge and the North
Entry (Figure 7a). The high copper concentrations in the nearshore area of Freda-Redridge are a
direct result of copper mine activity during the period 1895 ~ 1964 (Babcock and Spiroff, 1970;
Kerfoot et al., 1994). About 46 × 106 metric tons of stamp sands were dumped on the shore at
Freda-Redridge during this period. The original deposits of stamp sands have been reworked by
waves and wind-driven coastal currents and moved northeast toward the HN transect (Wright et
al., 1975; Wright et al., 1973). Comparison of the surface sediment copper distribution pattern
observed in this study with the pattern observed 25 years ago (Kraft, 1979) showed a general
similarity. However, concentrations of copper near the HN transect have increased, and the
copper has shifted toward the offshore regions. The coarse copper-rich sediments follow
bathymetric contours and reflect a general pathway of longshore sediment transport.
The wind-driven coastal Keweenaw Current, running northeastward along the shore,
appears mainly to be responsible for the longshore transport of fine sediments (Budd et al., 1999;
Hughes et al., 1970; Lien, 1973). The concentration of coarse sediments (Figure 7c) between
Redridge and North Entry suggested winnowing of fine fractions and transport much farther
along the shore and into deep basins. Depth profiles of particle size, copper concentration, and
mineral ratio (Albite:Chlorite) in the core taken near Copper Harbor (Figure 8) clearly showed
the long-term lateral movement of the clay- or silt-sized particles of copper-rich sediments.
Wave action was also a crucial factor contributing longshore transport of the coarse deposits.
Relatively heavy, coarse, copper-rich sediments were eventually moved northeastward and
deposited along the shore by waves. High concentrations of copper found along the shore
indicated longshore transport of the coarse sediments (Figure 7a,c). Bathymetry also could play
an important role in the longshore and cross-margin transports of the resuspendable coarse
sediments. Comparison of the copper distribution with the particle size distribution and the
bathymetry (Figure 7a,b,c) suggests that coarse copper-rich sediments are transported along the
shore until an abrupt change of bathymetry occurs near the HN transect. This observation
suggests that the coarse sediments resuspended by storms are moved to the deeper basin in the
North Entry area (Brassard and Morris, 1997; Hawley, 2000; Hawley et al., 1996; Rossmann and
Seibel, 1977). Finally, large drifting pack ice as a result of spring break-up of coastal ice also
may contribute to the longshore transport of the copper-rich sediments (Budd et al., 1999).
128 Jaebon g Jeong and S . Do u g las M c D owell Vol.2, No.2
Figure 7a
Figure 7b
Vol.2, No.2 Charact erizatio n and Tr ansport of Conta minated Sediment s in the Lake Superior 129
129
Figure 7c
0
5
10
15
20
25
0.0 1.0 2.03.0 4.0
[Cu], [C:N]/10 (molar), NMD (um), [Alb/Chl]
Depth (m)
[Cu] (mmol/g Sed
[C:N]/10, (molar)
Particle Size (NMD, um)
[Alb/Chl] ratio in XRD
Figure 8
130 Jaebon g Jeong and S . Do u g las M c D owell Vol.2, No.2
Although northeastward long-shore transport is a dominant process, cross-margin
transport also occurred in the study area as indicated by high copper concentrations in suspended
sediment along the HN transect (Figure 9). Cross-margin transport of the copper-rich coarse
deposits near the HN transect was caused by the transition in bathymetry in conjunction with
storms. The deposited copper-rich sediments in the coastal area are continuously re-suspended
by the waves in the high-energy environments and subjected to sorting with depth. As a result,
nearshore/offshore gradients in concentrations of copper in the surface sediments were found in
the study area. A mixing of the locally derived copper-rich sediments with lake sediments
resulted in concentrations of copper observed in the near- and offshore sediments (Figure 9)
which are lower than the typical values observed in the stamp sands from the Keweenaw
Peninsula (Jeong et al., 1999), but higher than Ontonagon sediments and Wisconsin red clay. The
copper concentrations of both of the latter are close to the typical values of soils and sediments
not contaminated by mining activity (Friedland et al., 1984; Ge et al., 2000; McLaren and
Crawford, 1973; Tessier et al., 1979). Despite the potential mobilization of copper in aquatic
systems, gradients or concentrations of total copper in sediments will provide an excellent tool
for investigation of the movements of the copper-rich particles and sediments in the study area
(Kerfoot et al., 1994; Kerfoot and Nriagu, 1999; Kerfoot et al., 1999; Mckee et al., 1989).
0.0
0.5
1.0
1.5
[Cu] (mg/g)
7.06 mg/g
Freda
Sources Settling Particles
(HN transect)Cores (Near
Copper Harbor)ON & WI
Sources
@ Cu Peak
@ Backgroud
Under Water Sample
@ Surface
0.0
0.5
1.0
1.5
[Cu] (mg/g)
7.06 mg/g
Freda
Sources Settling Particles
(HN transect)Cores (Near
Copper Harbor)ON & WI
Sources
@ Cu Peak
@ Backgroud
Under Water Sample
@ Surface
Figure 9
5. Conclusions
Metal-rich mine tailings are distinguished from the sediments from various non-mining
sources using a variety of techniques. The clearest differentiation between the tailings and
normal sediment sources, as well as differentiation among the sedimentary sources themselves,
came from a combination of bulk sample trace element determination using ICP or AAS
analyses, and X-ray diffraction analysis of the clay fraction. Copper, as expected, as well as Ag,
Co, and As were the most useful trace elements and were distinctly higher in the tailings relative
to all sediment sources. Simple X-ray intensity ratios from glycolated chlorite, illite, and
expandable illite/smectite peaks could distinguish among the mine tailings (chlorite rich),
sediment from the nearby Ontonagon River (low chlorite, high illite), and sediments from distant
Vol.2, No.2 Charact erizatio n and Tr ansport of Conta minated Sediment s in the Lake Superior 131
131
Northwest Wisconsin (low chlorite, high expandable phase). Major elements were not very
useful in distinguishing among the various sources, although the Wisconsin sediments were
unusually sodium- poor, and some differentiation could be made using mobile/immobile element
ratios. Analysis of coarser particles via SEM-EDS produced generally similar results for all
samples, except sediments from the Ontonagon River had distinctively higher P and lower S
concentrations. Electron microprobe analysis of particles was not informative. However relative
to the larger particles, the smaller particles tended to be higher in Al and S. Multivariate
statistical analysis of the particle chemistry also supported the uniqueness of the Ontonagon
River sediments.
Distribution of copper and signatures of minerals in nearshore sediments indicate that
metal-rich mine tailings are introduced into the coastal area. They moved into deep basins and
mixed with normal sediments in the lake via longshore and cross-margin transport. Longshore
and cross-margin transport of suspended particles and sediment in coastal areas are strongly
related to local environmental factors. Finally, particle size analysis was very important in
transport studies. Waves and bathymetry played major roles for longshore and cross-margin
transport of the coarse and heavy particles in the FR transect and near the HN Entry area.
Longshore transport of fine particles in conjunction with the Keweenaw current flowing
northeastward was predominant in the study area. In addition, vertical profiles of particle size,
copper, C:N ratios, and clay mineral ratios in the core sample taken near Copper Harbor gave
strong evidence of the long-range transport of the fine fraction of mine tailings. Thus, there
would be a high potential to contaminate the remote area of the lake by the metal-rich mine
tailings.
Acknowledgments
We would like to express our appreciation to the captain and crew of the R/V Laurentian
operated by University of Michigan and Blue Heron operated by University of Minnesota for
their assistance in field sampling. The research was supported in part by a fellowship from
Michigan Technological University and from the Keweenaw Interdisciplinary Transport
Experiment in Superior (KITES) project supported by National Science Foundation (NSF) award
number OCE-9712872.
References
Auer, M.T. and Gatzke, T.L., 2002. Cross margin transport and the timing of the spring runoff
event anf thermal bar formation in Lake Superior. Journal of Great Lakes Research in
press.
Babcock, L.L. and Spiroff, K., 1970. Recovery of copper from Michigan stamp sands: Vol. 1
Mine and mill origin, sampling and mineralogy of stamp sands, Inst. Mineral Res.,
Michigan Technoloogical University.
Bahnick, D.A., Markee, T.P., Anderson, C.A. and Roubal, R.K., 1978. Chemical Loadings to
southwestern Lake Superior from red clay erosion and resuspension. Journal of Great
Lakes Research 4(2), 186-193.
132 Jaebon g Jeong and S . Do u g las M c D owell Vol.2, No.2
Baker, J.E. and Eisenreich, S.J., 1989. PCBS and PAHS as tracers of particulate dynamics in
large lakes. Journal of Great Lakes Research 15(1), 84-103.
Brassard, P. and Morris, W., 1997. Resuspension and redistribution of sediments in Hamilton
Harbour. Journal of Great Lakes Research 23(1), 74-85.
Budd, J., Kerfoot, C.W., Pilant, A. and Jipping, L.M., 1999. The Keweenaw Current and ice
rafting: use of satellite imagery to investigate copper-rich particle dispersal. Journal of
Great Lakes Research 25(4), 642-662.
Chamley, H., 1989. Clay Sedimentology, Springer-Verlag Inc., New York.
Churchill, J.H., Ralph, E.A., Cates, A.M., Budd, J.W. and Urban, N.R., 2002. Observations of a
negatively buoyant river plume in a large lake. Journal of Great Lakes Research in press.
Colman, S.M. and Foster, D.S., 1994. A sediment budget for southern Lake Michigan: Source
and sink models for different time intervals. Journal of Great Lakes Research 20(1), 215-
228.
Cook, P.M., 1974. Asbestiform amphibole minerals: Detection and measurement of high
concentrations in municipal water supplies. Science 185, 853-855.
Cook, P.M., 1975. Semi-quantitative determination of asbestiform amphibole mineral
concentrations in Western Lake Sueprior water samples. In: W.L. Pickles, C.S. Barrett,
J.B. Newkirk and C.O. Ruud (Editors), Advances in X-ray Anlysis. Plenum Publishing
Co., New York, 557-567.
Cook, P.M. and Rubin, I.B., 1976. X-ray diffraction and electron beam analysis of asbestiform
minerals in Lake Supeiror waters. Annals no. 75CH1004-I 34-1, The Institute of
Electrical and Electronics Engineers, Inc., U.S.A.
Dassenakis, M., Degaita, A. and Scoullos, M., 1995. Trace metals in sediments of a
Mediterranean estuary affected by human activities. Sci. Total Environ. 168, 19-35.
Dell, C.I., 1973. A quantitative meneralogical examination of the clay-size fraction of Lake
Superior sediments. proc. 16th conf. Great Lakes Res., 413-420.
Forman, S.A. and Brydon, J.E., 1961. Clay mineralogy of Canadian soils. In: R.F. Legget
(Editor), Soils in Canada. University of Toronto Press, Toronto, 140-146.
Friedland, A.J., Johnson, A.H. and Siccama, T.G., 1984. Trace metal content of the forest floor
in the Green Mountains of Vermont: spatial and temporal patterns. Water Air Soil Pollut.
21, 161-170.
Ge, Y., Murray, P. and Hendershot, W.H., 2000. Trace metal speciation and bioavailability in
urban soils. Environmental Pollution 107, 137-144.
Vol.2, No.2 Charact erizatio n and Tr ansport of Conta minated Sediment s in the Lake Superior 133
133
Gutierrez, M.J.M., Achab, M., Sanchez, A., Morel, J.P. and Aguayo, F., 1996. Clay minerals in
recent sediments of the Cadiz Bay and their relationships with the adjacent lands and the
continental shelf. Advances in clay minerals.
Hawley, N., 2000. Sediment resuspension near the Keweenaw Peninsula, Lake Superior during
the fall and winter 1990-1991. Journal of Great Lakes Research 26(4), 495-505.
Hawley, N., Wang, X., Brownawell, B. and Flood, R., 1996. Resuspension of bottom sediments
in Lake Ontario during the unstratified period, 1992-1993. Journal of Great Lakes
Research 22(3), 707-721.
Heinrich, E.W., 1976. The Mineralogy of Michigan. Speaker-Hines and Thomas, Inc., Lansing,
Michigan.
Honeyman, B.D. and Santsch, P.H., 1989. A Brownian-pumping model for oceanic trace metal
scavenging: Evidence from Th isotopes. Journal of Marine Research 47, 951-992.
Hughes, J.D., Farrell, J.P. and Monahan, E.C., 1970. Drift-bottle study of the surface currents of
Lake Superior. Michigan Academician 2(4), 25-31.
Jeong, J., Urban, N.R. and Green, S.A., 1999. Release of copper from mine tailings on the
Keweenaw Peninsula. Journal of Great Lakes Research 25(4), 721-734.
Kemp, A.L.W., Dell, C.I. and Harper, N.S., 1978a. Sedimentation rates and a sediment budget
for Lake Superior. Journal of Great Lakes Research 4(3-4), 276-287.
Kemp, A.L.W., Williamson, J.D.H., Thomas, R.L. and Gregory, M.L., 1978b. Impact of man's
activities on the chemical composition of the sediment of Lake Superior and Huron.
Water, Air, And Soil Pollution 10, 381-402.
Kennedy, A.D., 1970. Recovery of copper from Michigan stamp sands: Vol. II Physical and
chemical properties of stamp sands. U.S. Bureau of Mines Project G0180241(SWD-18),
Institute of Mineral Research, Michnigan Tecnological University, Houghton.
Kerfoot, W.C., Lauster, G. and Robbins, J.A., 1994. Paleolimnological study of copper mining
around Lake Superior, Artificial varves from Portage Lake provide a high resolution
record. Limnology and Oceanography 39(3), 649-669.
Kerfoot, W.C. and Nriagu, J.O., 1999. Copper mining, copper cycling and mercury in the Lake
Superior ecosystem: An introduction. Journal of Great Lakes Research 25(4), 594-598.
Kerfoot, W.C. and Robbins, J.A., 1999. Nearshore regions of Lake Superior: Multi-element
signatures of mining discharges and a test of Pb-210 deposition under conditions of
variable sediment mass flux. Journal of Great Lakes Research 25(4), 697-720.
Kerfoot, W.C., Robbins, J.A., Harting, S. and Rossmann, R., 1999. Anthropogenic copper
inventories and mercury profiles from Lake Superior: Evidence for mining impacts.
Journal of Great Lakes Research 25(4), 663-682.
134 Jaebon g Jeong and S . Do u g las M c D owell Vol.2, No.2
Klump, J.V. et al., 1989. Variations in sediment acculation rates and the flux of labile organic
matter in eastern Lake Superior basins. Journal of Great Lakes Research 15(1), 104-122.
Kolak, J.J., Long, D.T., Beals, T.M., Eisenreich, S.J. and Swackhamer, D.L., 1998.
Anthropogenic inventories and historical and present accumulation rates of copper in
Great Lakes sediments. Applied Geochemistry 13, 59-75.
Kraft, K.J., 1979. Pontoporeia distribution along the Keweenaw shore of Lake Supeiror affected
by copper tailings. Journal of Great Lakes Research 5(1), 28-35.
Krezoski, J.R., 1989. Sediment reworking and transport in Eastern Lake Superior: In situ rare
earth element tracer studies. Journal of Great Lakes Research 15(1), 26-33.
Laberge, G.L., 1994. Geology of the Lake Superior region. Geoscience press Inc., Phoenix,
Arizona.
Lien, S.L., 1973. Circulation patterns in Lake Superior. Ph. D. Thesis, University of Wisconsin.
Lion, L.W., Altmann, R. and Leckie, J.O., 1982. Trace-metal adsorption characteristics of
estuarine particulate matter: evaluation of contributions of Fe/Mn oxide and organic
surface coatings", Environ. Sci. Technol., 16(10): 660-666. Environmental Science and
Technology 16(10), 660-666.
Martin, J.-M., Dai, M.-H. and Cauwet, G., 1995. Significance of colloids in the biogeochemical
cycling of organic carbon and trace metals in the Venice Lagoon (Italy). Limnology and
Oceanography 40(1), 119-131.
Mckee, J.D., Wilson, T.P., Long, D.T. and Owen, R.M., 1989. Pore water profiles and early
diagenesis of Mn, Cu, and Pb in sediments from large lakes. Journal of Great Lakes
Research 15(1), 68-83.
McLaren, R.M. and Crawford, D.V., 1973. Studies on soil copper, I. The fractionation of copper
in soils. Journal of Soil Science 24, 173-181.
Morel, F.M.M. and Hudson, J.M., 1985. Chapter 12. The geobiological cycle of trace elements in
aquatic system: Redfield revisited. In: W. Stumm (Editor), Chemical Processes in Lakes.
John Wiley & Sons, New York.
Nussmann, D.G., 1965. Trace elements in the sediments of Lake Superior. Ph. D. Thesis,
University of Michigan, Ann Arbor.
Ostrom, N.E., Long, D.T., Bell, E.M. and Beals, T., 1998. The origin and cycling of particulate
and sedimentary organic matter and nitrate in Lake Superior. Chemical Geology 152, 13-
28.
Reid, R.P., Pharo, C.H. and Barnes, W.C., 1980. Direct determination of Apatite in lake
sediments. Can. J. Fish. Aquat. Sci. 37, 640-646.
Vol.2, No.2 Charact erizatio n and Tr ansport of Conta minated Sediment s in the Lake Superior 135
135
Ribolzi, O., Valles, V., Gomez, L. and Voltz, M., 2002. Speciation and orgin of particulate
copper in runoff water from a Mediterranean vineyard catchment. Environmental
Pollution 117(2), 261-271.
Robertson, D.M., 1997. Regionalized loads of suspended sediment and phosphorus to Lake
Michigan and Superior-High flow and long-term average. Journal of Great Lakes
Research 23(4), 416-439.
Rossmann, R. and Seibel, E., 1977. Surficial sediment redistribution by wave energy: Element-
grain size relationships. Journal of Great Lakes Research 3(3-4), 258-262.
Sigg, L., 1985. Metal transfer mechanisms in lakes; the role of settling particles. In: W. Stumm
(Editor), Chemical Processes in Lakes. John Wiley & Sons, New York, 283-310.
Stumm, W. and Morgan, J.J., 1996. Aquatic Chemistry. John Wiely & Sons, New York.
Sunda, W.G. and Huntsman, S.A., 1995. Regulation of copper concentation in the oceanic
nutricline by phytoplankton uptake and regeneration cycles. Limnology and
Oceanography 40(1), 132-137.
Tessier, A., Campbell, P.G.C. and Bisson, M., 1979. Sequential extraction procedure for the
speciation of particle trace metals. Analytical Chemistry 51(7), 844-851.
Thomas, R.L. and Dell, C.I., 1978. Sediments of Lake Superior. Journal of Great Lakes Research
4(3-4), 264-275.
Thompson, M.E., 1978. Major ion loadings to Lake Superior. J. Great Lake Res. 4(3-4), 361-
369.
Van Luven, M.D., Huntoon, J.E. and Maclean, A.L., 1999. Determination of the influence of
wind on the Keweenaw Current in the Lake SUperior basin as identified by Advanced
Very High Resolution Radiomenter (AVHRR) imagery. Journal of Great Lakes Research
25(4), 625-641.
Whitfield, M. and Turner, D.R., 1987. Chapter 17. The role of particles in regulating the
composition of seawater. In: W. Stumn (Editor), Aquatic Surface Chemistry. Hohn Wiley
& Sons, New York, 457-493.
Wright, T.D., Hamil, B.M., Kraft, K.J., Leddy, D.G. and Nordeng, S.C., 1975. Analysis report-
Keweenaw waterways maintenance dredging. DACW 34-74-C-0149, U.S. Army Corps
of Engineers, St. Paul District.
Wright, T.D., Leddy, D.G., Kraft, K.J. and Spain, J.D., 1973. Environmental impact assessment
for Keweenaw Waterway Michigan. National Biocentrics Inc., St. Paul, Minn., pp. 1-132.