“i

§3

(k)

M

<33

f0

('1

(\3

This is to certify that the

thesis entitled

Assessment of Zinc Accumulation
In Sediments of the Great Lakes

presented by

Jeffrey Byrem Vought

has been accepted towards fulfillment

Master's

of the requirements for

 

Date 3'4 JUL: 0 i

0-7639

 

degree in Mural
Geosciences

MM] ’7

Major professor

     

MS U is an Ajfirmau've Action/Equal Opportunity Institution

 

 

 

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c-JCIRC/DatoDuo.pB§-p.15

 

ASSESSMENT OF ZINC ACCUMULATION
IN SEDIMENTS OF THE GREAT LAKES

By

Jeffrey Byrem Vought

A THESIS
Submitted to
Michigan State University

in partial fulfillment of the requirements

for the degree of

MASTER OF SCIENCE

Department of Geological Sciences

2001

ABSTRACT

ASSESSMENT OF ZINC ACCUMULATION IN SEDIMENTS OF THE
GREAT LAKES

By

Jeffrey Byrem Vought

Sediment cores taken from lakes Superior, Michigan and Ontario
were used to examine spatial and temporal variations in anthropogenic zinc
loading through calculation of zinc inventories and accumulation rates.
Inventories and accumulation rates show anthropogenic loading increases in
the order of: Lake Superior<Lake Michigan<Lake Ontario. Trends in zinc
accumulation rates generally show loading increases from 1900 to 1970
followed by a decrease to the present. Atmospheric contributions of zinc
loading to the lakes are Lake Superior (100%), Lake Michigan (65%) and
Lake Ontario (20%). Trends in zinc accumulation rates are related to
production and consumption histories, emplacement of environmental
regulations or a combination of both and are similar to cadmium and
mercury implying that these metals may have similar environmental

controls.

To Mom, Cynth and LuWanna for years of support.

iii

ACKNOWLEDGMENTS

I would like to thank Jon Kolak and Gary Icopini for their support
and information they were always willing to provide and also for showing
me what it took to really produce reliable data and analyze it accordingly. I
would especially like to thank Dave Long because without him this study
and its remarkable context would have never been possible. Also many
thanks to the National Oceanic and Atmospheric Administration for funding
and also to the International Association of Great Lakes Research for

providing data essential to this study.

iv

TABLE OF CONTENTS

LIST OF TABLES vi
LIST OF FIGURES vii
I. INTRODUCTION 1
II. METHODS AND PROCEDURES 5
Field Methods 5
Laboratory Methods 7
Integrity of Zinc Profiles 9
me Dating and Sedimentation Models 13
III. RESULTS AND DISCUSSION 17
Concepts 17
Anthropogenic Zinc Inventories and Sediment Burdens 19
Recent Accumulation Rates of Zinc 29
Historical Accumulation Rates of Zinc 34
Atmospheric Deposition of Zinc 37
Possible Sources of Zinc 43
Comparison With Other Contaminants in the Great Lakes 51
IV. CONCLUSIONS 56

REFERENCES 59

Table 1

Table 2

Table 3

Table 4

LIST OF TABLES

Percent atmospheric contribution of zinc relative to
total inputs.

Productive and consumptive uses of zinc as possible
source functions.

Comparison of zinc accumulation characteristics
with those of cadmium and mercury.

Environmental summary of zinc in the Great Lakes.

vi

4O

46

53

57

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

LIST OF FIGURES

Sediment and IADN atmospheric sampling locations.

Vertical profiles of: a) zinc b) erosion and c) arsenic.

Relationship of (a) background concentration and

background depth to vertical concentration changes

of Pb in sediments and (b) changes in sediment Pb

concentration as a function of time (Long et al., 1995).

Histogram of uncorrected and focusing-corrected

anthropogenic zinc inventories.

Zinc sediment accumulation rates.

Temporal variations of zinc accumulation rates
from lakes Ontario (LO4OA), Michigan (LM68K)
and Superior (NOAA3).

Zinc sediment accumulation rates at representative

sites in the Great Lakes compared to

production/consumption data.

vii

'11

18

23

31

35

47

INTRODUCTION

Increasing concerns of quality degradation of the Great Lakes
ecosystem has led to many studies of anthropogenic accumulation of trace
metals including zinc (Sweet et al., 1998; Johnson and Nicholls, 1988; Hart,
1982; Kemp et al., 1978a). The International Joint Commission (US. and
Canada) has grouped metal contaminants in the Great Lakes based on their
potential of being an environmental hazard with the lowest level
representing the highest priority. Mercury (Hg) and lead (Pb) are level 1,
arsenic (As), cadmium (Cd) and selenium (Se) are level 2, and copper (Cu),
zinc (Zn), chromium (Cr) and vanadium (Vd) are level 3 (International Joint
Commission, 1978). Level 1 metals are most often the focus of studies in
' the Great Lakes due to their high toxicity as compared to level 2 and level 3
metals. Level 2 and 3 metals such as zinc may become toxic, however,
when the loading capacity of the sediment is exceeded and there are no
longer any sediment reservoirs to sequester the increased anthropogenic
burden (Stigliani et al., 1991). Principal pathways of metal input to the
Great Lakes are: (1) overland flow (2) ground-water (3) diagenetic

remobilization and (4) atmospheric deposition (Kelley et al., 1991; McKee

et al., 1989a; Van Loon and Beamish, 1977). Atmospheric deposition can
contribute up to 100% of metal loading to the Great Lakes (Long et al.,
1995), except for near coastal areas influenced by runoff, river or sewage
outflows (Sweet et al., 1998; Hoff, 1994; Edgington and Robbins, 1976).
Upon entering the lakes, metal contaminants are removed from the water
column by adsorbing onto suspended particulate matter (Hart, 1982). The
rapid settling velocities and hence short residence times of particulate
matter in the water column compared to long hydraulic residence times of
water in the lakes enable the loading history of contaminants to be
preserved in sediment columns (Golden et al., 1993; Schmidt and Andren,
1984)

Numerous studies have examined the source, transport, and fate of
contaminants in the Great Lakes region using sediment cores (Kolak et al.,
in press; Jeong, 1994; Eisenreich et al., 1986; Tessier et al., 1980).
Examples of contaminants studied are PCB and DDT (Golden et al., 1993),
lead, mercury, arsenic and cadmium (Long et al., 1995; Johnson and
Nicholls, 1988; Eisenreich et al., 1986), chromium (Jeong, 1994) and
copper (Kolak et al., in press). These studies have examined the spatial

distribution of total anthropogenic loading (Jeong, 1994; Johnson and

Nicholls, 1988; Kemp et al., 1978a), the history and distribution of
anthropogenic accumulation (Long et al., 1995; Edgington and Robbins,
1976), the dominant pathways of contaminant transportation (Sweet et al.,
1998; Kelley et al., 1991; Christensen and Osuna, 1989; Van Loon and
Beamish, 1977) and possible sources (Wong et al., 1995; Eisenreich, 1980;
Winchester and Nifong, 1971). These studies generally found increases in
trace metal loadings to the Great Lakes from the late 1800’s to the late
1960’s corresponding with urbanization and industrialization of the region
(Long et al., 1995; Nriagu et al., 1983; Kemp et al., l978a). Decreases in
trace metal inputs from the late 1960’s to the present were attributed to
changes in production and consumption of the metals and the enactment of
environmental regulations focused on these uses (e.g. the restriction of
gasoline containing Pb) (Long et al., 1995; Eisenreich, 1992). Previous
studies found metal input to Lake Superior dominated by the atmosphere
(Eisenreich et al, 1992), input to Lake Michigan dominated by atmospheric
fallout from Chicago and Gary (Eisenreich, 1980; Winchester and Nifong,
1971), and input to Lake Ontario dominated by effluents from Hamilton

Harbour and the Niagara River (Nriagu et al., 1996; Mudroch, 1983).

Spatial and temporal distribution of zinc accumulation in the Great
Lakes on a regional scale is not yet fully understood. Previous studies are
limited by their smaller scale because they only define point sources of the
contaminant (Johnson and Nicholls, 1988; Tessier et al., 1980) or examine
lake specific trends in contaminant deposition (Kelley et al., 1991; Kemp et
al., 1978a). Therefore, the goals of this study are to regionally: (l) quantify
the amount of anthropogenic zinc loading by determining anthropogenic
inventories and sediment burdens (2) examine the spatial variation in these
inventories (3) examine temporal trends of zinc accumulation rates (4)
examine the spatial distribution of recent accumulation rates (5) estimate the
relative proportion of atmospheric inputs (6) evaluate possible sources of
zinc by comparing loading histories with various production and
consumption histories (7) identify possible controls of zinc accumulation
rates and (8) compare zinc accumulation characteristics with those of other

contaminants.

METHODS AND PROCEDURES

Field Methods

Box cores of sediment were collected from sites in lakes Michigan,
Superior and Ontario from 1990 to 1992 aboard the R\V Lake Guardian
(USEPA), the R\V Seward-Johnson and the R\S Johnson Sea Link [I
(NOAA-NURC). Sites were located to maximize sampling area so that
local and regional depositional gradients could be examined and also
located within depositional basins because when studying metal loading it is
necessary to study zones of active sediment accumulation (Figure 1). To
ensure sampling integrity, only undisturbed samples were used which were
determined by me profiles and presence of sediment-free water above a

level sediment surface in the box core.

The box cores, once aboard ship, were sub-cored for 210Pb, trace
metal and organic contaminants using 7 cm diameter polycarbonate tubes.
The tubes were inserted into the box cores under vacuum to avoid sediment
compaction and 5 cm away from the box core walls to prevent sample
contamination. A hydraulic extruder was used to section the cores into

0.25, 0.5, 1.0 and 2.0 cm increments. Smaller increments were used near

.3232: wEEES 05.3883 22 one EBB—com A onE

:33

     

ESE

 

ave

omm
SSEQ 83% ‘23
N8 8‘3

:3». as: £3

3% wEEEmm 75¢: O
3% 0335858 834 I

8:» wfiafiwm 1550 O

the tops of the cores to obtain the highest resolution needed to examine
recent changes in loading. A Teflon® spatula was used to remove the outer
surfaces of the cores to minimize smearing effects from sub-core walls.
Sub-core sections used for trace metal analysis were placed in
polypropylene tubes previously acid-washed and rinsed with distilled-

deionized water (DDW) and frozen until sample analysis.

Laboratory Methods

Samples were thawed at room temperature and homogenized with a
Teflon® spatula. Five grams of each sample were weighed and dried in a
convection oven for 24 hours at 50°C. The weights were recorded before
and after drying to determine porosity. Dried samples were ground into a
fine powder using a ceramic mortar and pestle, and digested using a nitric
acid microwave digestion technique to determine the total metal content of
the hydromorphic fi'action (Hewitt and Reynolds, 1990). InstraTM grade
ultrapure HNO3 (15M) was used to minimize sample contamination. The
resulting leachates were diluted by adding 90 ml of DDW, filtered through

.40um acid cleaned NucleporeTM membrane filters and stored at 4°C in

polypropylene sample bottles until analysis. Leachates were analyzed for
zinc by flame atomic absorption spectrophotometry on a Perkin Elmer
Zeeman 5100.

To ensure quality control, ultra-clean procedures were used during all
phases of analysis. Sediments were only allowed contact with plastic and/or
Teflon® surfaces that had been previously acid-washed. Sampling
equipment was cleaned by washing in 10% HCL at 20°C for 24 hours,
rinsing and soaking in DDW for another 24 hours, and drying under a hood
supplied with Class 100 filtered air.

The accuracy of the digestion technique was determined using a
standard reference material (SRM) (NTST Buffalo River Sediment #2704).
Acceptable sediment extracts had metal recoveries (R) in the range
70%<R<120% of the certified value of the SRM. Precision of the digestion
was monitored using subsplit samples (5% of all samples) where relative
standard deviations (rsd’s) of the samples were required to be less than
15%. Analysis precision was assured using replicate analyses on particular
samples (10% of all samples) where acceptable results had rsd’s also less
than 15%. Sample blanks were analyzed with every sediment digestion and

considered unacceptable if they exceeded the method detection limit

(MDL). The MDL was defined as three times the standard deviation of
seven replicate analyses of a representative blank (Glaser et al., 1981) and
was equal to 6 mg/L.

Methods for me dating are based upon the work of Robbins and
Edgington (1975) and were modified according to the work of Eakins and
Morrison (1977). 2l"Pb concentrations were determined through the
analyses of 210P0 (daughter of me). Sediments used for dating were stored
and refrigerated in plastic jars. Sediment samples were dried, spiked with a
known amount of 208P0, ground into a fine powder and digested using
concentrated HCl. The solution was transferred to a silver plating cell and
inserted into an alpha spectrometer. 210Po activity was calculated as the
ratio of the 210P0 to the added 208Po spike. 210Po activity was used to
determine the unsupported 21"Pb concentration. Age was determined as the
ratio of unsupported 2|"Pb activity relative to that at the surface and it’s half-

life.

Integrity of Zinc Profiles
When using concentration profiles to evaluate loading histories, the

possibility of post depositional alteration of sediments (e. g. physical

reworking, diagenesis and bioturbation) must be considered (McKee et al.,
1989a). Post-depositional processes such as current erosion, slumping and
sliding can alter concentration profiles in sediments. Site LO4OA in Lake
Ontario has a typical zinc profile for sediments in the Great Lakes (Figure
2a) where as depth decreases there is an increase from background
concentration to peak concentration followed by a decrease near the top of
the core. Site LM23 in Lake Michigan is an example of an erosional site
evidenced by the 21"Pb dates and the location of the peak zinc concentration
at the sediment water interface where the upper portion of the profile that
recorded the recent decrease has been removed by current erosion (Figure
2b).

Upon entering the lakes, the majority of zinc is sequestered by
organic matter and the remaining portions by Fe/Mn oxides and clay
minerals (McKee et al., 1989b; Cahill and Shimp, 1984; Forstner and
Wittman, 1979). Early diagenesis, the post-depositional chemical changes
of these phases which occur in the porewater, can alter the concentration
profiles of trace metals (Bemer, 1980). Early diagenesis is caused by the
depletion of oxygen levels in the pore water because of utilization by

microbes during the consumption of organic matter (Atlas and Bartha,

10

638.8 G was .8696 3 0:? Q «o $an 133.5 > .N oSwE

3R1: 3:295. AwhéoEN 3B3 oEN

em 8 S S w e o 2: c2 2; 8L 2: 8 8 25 8m 8w 8m 8m 2:

 

 

 

 

 

 

 

LHJJIjlnlq-LJitaJIlLILI-Jii-Lnjdiqq14.1J_m—l LLLLL_LLLL LL L4 L._._L:LL-TL.NMI 3.4L.-_;JI-ijoml
. -_ _ L r
_ L _ L L
. _ ._ L L L
- NTL. a L .m - Lov-
L _ L- “ LVN- L L
.- .m L L L L- L MR-
L L8- L ,
_ L . r
w. _, . ,2- L Amm-
L III-III I. ,. L L L. L
,_ T-tI-IIIMIIII _ . - L .
.- --liI-I-i-II . o.- v - It I
4‘ w L LNT ,- LON
_, ._ L x .2-
L L a- W \K- M
L- L _m- _ L A A:-
- .. . I v. L
L L \\\H L L- mm-
L A _- \Ii\\\-o\\ r L
LL_LPLLLLLLLLLLL.p__L___p_L»r o Fry-FL HO fi__L_IL~rFLL»LLLLr» LLLLLL o

m<<OZ cam 824 8.5 <30; 8mm

(we) laden

11

1993). As organic matter is consumed, the metals it contains are released
into the porewater due to the changes in the oxidation-reduction (redox)
conditions. The metals are then either adsorbed by other sediment phases or
are released into the overlying water column (McKee et al., 1989b). The
result of early diagenesis is that concentration profiles may not accurately
reflect the loading history of the contaminant.

Many studies have found that diagenesis only has a slight impact on
total zinc concentration profiles in sediments. McKee (1989b) found
diagenetic remobilization of zinc is only evident when studying individual
hydromorphic phases of sediment and also found that total metal profiles,
such as those used in this study, are indicative of anthropogenic inputs
because the mass of zinc mobilized by early diagenesis is insignificant
compared to total concentrations. The comparison of contaminant profiles
controlled by diagenesis with contaminant profiles where the role of
diagenesis is unknown has also been used to suggest if this process is a
factor (Kolak et al., in press). Arsenic profiles, strongly influenced by early
diagenetic processes, have large peaks located at the redox boundary due to
the upward mobilization of arsenic under reducing conditions and

precipitation upon entering oxidizing conditions (Long et al., 1995; Moore

12

et al., 1988). The comparison of arsenic profiles with zinc profiles at site
NOAA3 (Figure 2c) show zinc profiles do not have a peak at the redox
boundary suggesting it is relatively unaffected by diagenesis and also
supporting the use of sediment cores to examine loading histories of the

metal.

””Pb Dating and Sedimentation Models

- To assess contaminant loading histories, sedimentation rates and
sediment ages can be determined through analysis of ”on profiles in
sediments (Wong et al., 1995; Golden et al., 1993; Eisenreich et al., 1986;
Robbins, 1978). Sources of ”on in sediments are atmospheric fallout of
”on and in-situ radioactive decay of 226R3. Atmospheric ”on, also called
unsupported ”(’Pb, originates from the decay of the noble gas 222Rn which is
a daughter product of uranium and has a high affinity for atmospheric
particulate matter (Robbins, 1975). Once ”on enters the water column via
wet and dry deposition of atmospheric particulates, it is rapidly transferred
to sediments where there is a source of 226Ra and therefore decays towards
these supported or in-situ levels. Sediment profiles of”on often have a

region of constant activity in surface sediments caused from mixing by

13

oligochaete worms (Christensen and Chien, 1981) followed by an
exponential decrease below attributed to ”on decay (Wong et al., 1995;
Schell, 1986; Robbins, 1982). Sediment ages were determined using
profiles of ”on activity and the known decay rate. These profiles can be
modeled using a rapid steady-state mixing model with a constant
sedimentation rate (Golden et al., 1993). In this model new material added
to the sediment surface is mixed instantaneously and homogeneously
throughout a mixed zone of depth S(cm). This mixed zone migrates upward
at a rate equal to the sedimentation rate W(g/cm2yr). Sediment below the
mixing zone is considered unaffected by bioturbation. For a constant ”on
flux, the mean sedimentation rate (W) can be determined using the

equations (Robbins, 1978):

A(z)=Am forzSS (1)

A(z) = Am exp{[-k(z -S)]/W} for z > S (2)

where A(z) equals the unsupported ”on activity at mass depth 2, Am is the

unsupported 210Pb activity in the mixed zone, 2 is the mass depth (g/cmz)

l4

and k is the ”on decay constant (0.0311 yr“). Sediment compaction effects
are removed if S is expressed as a cumulative dry mass (g/cmz) and W is
expressed as an accumulation rate (g/cmzyr) (Wong et al., 1995). If there is
no mixing zone, the rapid steady state mixing model simplifies to a constant

flux-constant sedimentation (cfzcs) model (Robbins, 1978):

A(Z) = A5 eXP(-kz/W) (3)

where AS is the unsupported 210Pb activity at the sediment water interface.

Sedimentation rates were derived through the examination of linear
regressions of unsupported ”on versus cumulative dry mass (CDM) and
agree with sedimentation rates previously determined (Eisenreich et al.,
1989; Kemp et al., 1974). Processes such as slumping, erosion and
bioturbation which cause changes in sedimentation rates can also be shown
using these linear regressions. Core sites SJEII, DTL and LMll for
example, may contain slump features making them unable to be used for
examination of historical zinc inventories (Kolak et al., in press). The
majority of sediment ages were determined using ”on concentrations and a

constant flux-constant sedimentation (cfzcs) model because this model

15

yielded dates which agreed better with the production histories of
contaminants compared to the model using a no-mixing constant
sedimentation rate (crs) (Wong et al., 1995).

To assess the impact of bioturbation on contaminant profiles, the
residence time of the sediment within the mixed zone must be considered
(Kolak et al., in press; Eisenreich et al., 1989; Robbins, 1982). The
residence time, or intrinsic resolution, is the length of time over which a
change in the input rate of a tracer will not be observed in the sediment
record (Eisenreich et al., 1989). The intrinsic resolution of a core can be
estimated as S/W. The result of intrinsic resolution on profile analysis is
that two events. become unresolvable when the time separating them is less
than the residence time of the sediment in the mixing zone (Wong et al.,
1995; Robbins, 1982). In the Great Lakes, intrinsic resolutions are on the
time scale of decades (Wong et al., 1995; Eisenreich et al., 1989).
Therefore when examining zinc profiles, long-term patterns in zinc
deposition (e. g. longer than a decade) will remain and annual patterns will

be erased due to bioturbation within the mixing zone.

16

RESULTS AND DISCUSSION

Concepts

To calculate anthropogenic zinc loading to the Great Lakes, the
natural contribution (e. g. from mineral weathering) must be first
ascertained. This natural or background input is assumed to remain
relatively constant through time. The anthropogenic contribution is defined
as the difference between the total and background concentration. The
background depth is the depth in the profile below which the zinc
concentration remains relatively constant and the background concentration
is the average concentration below this depth. Figures 3a and 3b depict a
typical zinc profile, the definitions of background depth and concentration,
and a typical plot of Pb concentration versus sediment age. All cores except
G32 and E30 reached background depth which is similar to other studies
utilizing the same cores but focusing on elements such as Cu, Pb, As, Cd,
Hg, and Cr (Kolak et al., in press; Long et al., 1995; J eong, 1994). Average
pre-industrial background concentrations for lakes Michigan, Superior,
Ontario and the entire region (the average of all three lakes) respectively are

85, 93, 130 and 103 ug/g dry sediment weight. The background

17

 

 

L L . 1 4 L i 1 t L t 4 A L l . . LI

L a U8:
_- “om:
L :080 m0 50> /’ H L

.- -H 82
fl- GOflNbH—QOGOU £60m Omwfi
L .82
L Toma
writ-L.-- II--. , . _ . . In 88

 

 

o2 ONL om ow o

885:5 :0:m3-::08€0m

8R3 am 5:883:00

 

A33 in :0 god 2:: :0 5:05.: 0 m:
5:888:00 am 80868 E memes-.5 3v :5 3:08:03 5 pm :0 mamas-L0 5:888:00
30:5» 0: 5:0: “058853 :5 5:888:00 “058303 A8 :0 :Em:0:m_0m .m 0.53:

 

 

 

 

 

 

 

 

 

L: ,;;-.Jom

L .2.
5:888:00 .
:::0._wo_0:m .
3 Y .
e . .
l— L -

.- 595 5:89—23 om

.- I .l III-IIIIW-m .ofi
L III-II“ act-emu M
I l- .

I _ _ c L P c _ L o

 

 

 

AN

.IIIImEE-L-

o-S ow ow

8%.: am 5:888:00

(we) mdad

18

concentrations for lakes Superior and Ontario are comparable to and those
for Lake Michigan are higher than values reported in a previous study
(Mudroch et al., 1988). These background concentrations, however, do
compare favorably with values reported for uncontaminated soils in the
upper Midwest region (Shacklette and Boemgen, 1984; F 6rstner and
Wittman, 1981). Background concentrations of zinc within the same lakes
have variations up to 73% possibly caused by differences in sediment
composition (e. g. amount of organic material, mineralogy and fine-grained

material) or post-depositional core disturbance (Long et al., 1995).

Anthropogenic Zinc Inventories and Sediment Burdens

To examine the spatial distribution of contaminant loading and the
pathways of delivery to the Great Lakes, contaminant inventories must be
examined (Long et al., 1995; Wong et al., 1995; Jeong, 1994; Golden et al.,
1993; Eisenreich et al., 1989). The contaminant inventory represents the
total mass of the metal accumulated from anthropogenic inputs. The
inventory is vertically integrated over the entire sediment core and can be
defined as the concentration of zinc per unit area. For inventories to

represent only anthropogenic inputs, the background zinc concentration is

19

subtracted from the total concentration. The equation used for inventory

calculation is:

Ian} (pg/ems = 2:. [C.sed' x {(1-¢> x p x cm (4)

where i = metal of interest, j = site for the sediment core, x = number of

depth increment, Cised' = background corrected metal concentration in the

sediment (ug/ g dry weight), (I) = sediment porosity, p = sediment dry density

(g/cmz), d = thickness of the sediment increment (cm) and Invji = inventory

(ug/cmz) of the ith metal in the sediment core at the j‘h site. To compare
inventories between different sites, variations in sediment accumulation
rates due to sediment focusing must be considered. Sediment focusing is
the movement of recently deposited sediments from various shallow areas
of the lake to deep, quiescent depositional zones (Hilton et al., 1986).
Variations of inventories caused by sediment focusing can be corrected by

normalization through the use focusing factors in the equation:

CInvji (pg/cmz) = Invji / FF] (5)

20

where i = metal of interest, j = site for the sediment core, F F J = focusing

factor at the j‘l‘ site and Clnvii = corrected inventory (ug/cm") for the ith

metal in the sediment core at the jth site. The FF value is equal to the ratio
of the measured 21"Pb to the expected 2“’Pb from atmospheric deposition.
The expected 210Pb is the of amount of 2"’Pb in peat bogs which are used as
a reference for historical atmospheric deposition because these reservoirs
only have an atmospheric source of 21"Pb (a value of 15.5 pCi/cm2 was used;
Golden et al., 1993). This value is considered constant because the
atmospheric flux of 210Pb is relatively the same throughout the Great Lakes
region (Robbins etal., 1975).

Intralake inventory comparisons, the comparisons of inventories at
different sampling sites within the same lake, indicate lake specific trends of
contaminant loading. Interlake inventory comparisons, the comparisons of
sampling sites between different lakes, can suggest whether or not there are
regional trends in contaminant loading. For example, Long et a1. (1995)
used similar interlake inventories of Pb and Cd to suggest that atmospheric
deposition plays an important role in the loading of these metals to the
entire Great Lakes region. The same study, however, also used similar

intralake inventories to imply that the lakes were historically well-mixed

21

with respect to the contaminants of interest. In examining these
explanations it is important to note that similar intralake inventories can be
caused by the dominance of a uniformly loading input (e. g. the atmosphere),
the presence of well-developed mixing, or both processes acting together.
Similar interlake inventories, however, suggest the presence of a regional
uniformly loading input. This regional input strongly implicates the
atmosphere because of the large spatial scale of the Great Lakes and the
corresponding lack of uniformly loading pathways that operate on this

scale.

Uncorrected and corrected inventories for each site are shown in
Figure 4. Focusing-corrected zinc inventories generally decrease in the
order of Lake Ontario, Lake Michigan and Lake Superior, respectively.
Mean corrected inventories for lakes Superior, Michigan and Ontario are
48, 195 and 693 ug/cmz, respectively. In lakes Michigan and Ontario,
intralake comparisons of uncorrected zinc inventories yield significant
variation because of spatial differences in sediment focusing. This variation
is significantly reduced when the inventories are corrected using focusing
factors. Lake Ontario sediments have received anthropogenic contributions

of zinc at significantly higher concentrations than Lakes Superior and

22

02.6802: 0:8 0808:0880 “0080930500.: 0:: 3:08.505 .8 8083mm.“ .v 0:03.:

:0:::~> .8 8005000 3:09:00 n >00 :0:0:0> .8 8065000 3:00:85 u >00

 

 

 

 

 

 

 

 

 

 

 

 

 

 

    

   

 

 

 

 

 

 

 

 

 

 

 

 

 

l l

l

lllllll

00:00:38 00:00:00 H, - I A a 008808: 388:0
a I u o o a
>> m m A V z m. A

m8 omm <o§ 20a :3: 0.24833 m2. v80 22. 8.2 92 9202 EB A5

1 n

I \m “A“.
.[I r IN. [I
.I. r. H H :0:
I ll :8: 83900 l .3835
I .X.w~fl>00 II Av\..Om..u>00 I $m~n>00
I {gov ">03 1I fem: n >03 I $3 ">03
1 I I .. 8:08
- 83:0 -- £0208 . . m

 

 

 

 

 

o
W.
1..
8m 0
(M
)
a. m
0
w
z
(
82

23

Michigan which suggest that this lake is relatively more influenced by
localized sources.

Reductions in variation can be quantified through the examination of
the coefficients of variation (CV) for the uncorrected and corrected
inventories within each lake (Kolak et al., in press; Wong et al., 1995)
(Figure 4). Cores LMl 1, SJE2 and DTL were not included in these
calculations because they may be disturbed as indicated previously using
210Pb profiles. Significant reductions in coefficients of variation of
corrected inventories indicate that accounting for sediment focusing
significantly reduces the differences in zinc inventories for all three lakes.
This reduction in variability when observed in studies of other contaminants
has been used to imply that historical loading to the lakes was either
dominated by one source or was well-mixed with respect to the contaminant
of study (Wong et al., 1995). The large CV that remains for Lake Michigan
after correcting for sediment focusing compared to the CV of lakes Superior
and Ontario suggests that historically this lake is relatively not well-mixed
with respect to zinc. Characteristics attributed by the comparison of CV’s
are limited, however, because they are only generalizations of lake-wide

behavior.

24

A more specific method of assessing variation in trends of corrected
inventories is direct comparison of CInv’s among various sampling sites.
To compare CInv’s between sites it is necessary to determine the
uncertainty associated with CInv’s at each site. Uncertainties were
determined through the propagation of errors from contributing parameters
used in the calculation of CInv. Uncertainties of each CInv were calculated
using the equation (from Kolak et al., in press; modified from Bevington,

1969)

— *
049qu " —In_v + “—5: ' .2._anvFF (7)

CInv2 Inv2 F F2 Inv*FF

where 0' represents the standard deviation of the indicated parameter. The
third term on the right hand side of Equation (7) represents the covariance
of Inv with FF. If there is no correlation between Inv and FF this term
reduces to zero and is eliminated from the equation (Kolak et al., in press).
A linear regression of Inv versus FF indicates there is no correlation

between the two parameters (r2 = .23) and equation (7) can be rewritten as:

25

cam, = +/- (CInv2 * (c,,,,2 /1nv2 + (fig/12132))“2 (8)

Uncertainties for each site are calculated using Equation (8) and are plotted
in Figure 4. Relatively similar intralake inventories in Lake Ontario imply
it may be historically well-mixed with respect to anthropogenic zinc inputs,
may be locally dominated by atmospheric deposition or a combination of
both. Likewise, significant differences in intralake inventories in lakes
Michigan and Superior suggest that zinc inputs to these lakes are either not
historically well-mixed, are dominated by localized sources or both. Large
variations in interlake inventories either shows that atmospheric deposition
cannot be the dominant input of zinc for the Great Lakes region or that if it
is dominant, it is in the form of a regional depositional gradient.

Corrected zinc inventories in the southern basin of Lake Michigan are
slightly higher than inventories in the northern basin. The higher
inventories may be related to the proximity of these sites to the large
urban/industrial center around southern Lake Michigan (Sweet et al., 1998;
Eisenreich, 1980; Winchester and Nifong, 1971). The lower inventory of
site 70M may be caused by a bathymetric barrier which isolates this site

from sediment focusing processes affecting other sites in the lake (Simcik et

26

al., 1996). In Lake Superior corrected zinc inventories are higher in the
Duluth Basin compared to other intralake sites. The sites in the Duluth
Basin (DTL and SJE2) may be disturbed as indicated by 21"Pb profiles
(Long et al., 1995). If these sites are not disturbed, however, the spatial
distribution of inventories in the basin may reflect the discharge of zinc
from the Duluth River. Site DTL has a higher inventory and is closest to
the tributary compared to SJE2 which is expected if the river is a source of
the metal.

The absence of anthropogenic zinc loading to site 1383 in Lake
Superior is quite interesting. The same trends exhibited by chromium in the
Duluth Basin were attributed to dilution by taconite tailings even though the
metal concentrations in these tailings are unknown (J eong, 1994; Andrew,
1970). There is also the possibility of an anthropogenic input that is
significantly lower than the range of background concentrations. Kemp et
al. (1978b) found that red clay bluffs on the Wisconsin shoreline account
for up to 58% of the total fine-grained input to Lake Superior and have a
low organic matter content. Lake Ontario, however, has shown a two to
threefold increase in organic matter in the past one hundred years (Kemp et

al., 1974). These levels of organic matter are reflective of the ability to

27

sequester zinc and may contribute to the relative lack of anthropogenic
input to Lake Superior.
Using corrected zinc inventories it is possible to calculate the

anthropogenic sediment burden (ASB) of the metal for each lake:

ASB'i (kg) = ((ZCInviiy n1) x SA, (6)

where i = metal of interest, 1 = lake, ASBli= anthropogenic sediment burden
for the ith metal in the 1th lake (kg), CInvii = corrected inventory (uglcmz) for

the ith metal in the 1th lake, n; = number of sites and SA. = surface area of the
1th lake (m2).

Anthropogenic sediment burdens for Lake Superior, Michigan and
Ontario are 5.64 x 107 kg, 1.98 x 108 kg and 3.39 x 108 kg, respectively.
Intralake comparisons of zinc ASB’s and inventories indicate that Lake
Superior is least impacted by anthropogenic loading and Lake Ontario is the
most impacted by anthropogenic loading. This is expected considering the
proximity of these lakes to major sources of zinc where Lake Ontario is
located closest to the high concentrations of urban/industrial activities

relative to Lake Superior.

28

Recent Accumulation Rates of Zinc

Recent temporal trends of zinc accumulation in sediments are
analyzed through the calculation of sediment accumulation rates (SAR’s) of
the metal. The SAR for each sediment increment is calculated using the

equation:

SAR‘j (g/mzyr) = c‘j x wj x 10-2 (9)

where i = metal of interest, j = site for sediment core, C3 = background

corrected concentration (ug/ g) of ith metal in the surficial sediment at the jth
site, W, = mass sedimentation rate (g/cmzyr) in the jth site and SARij =

sediment accumulation rate (g/mzyr) for the i‘h metal at the j‘h site. To
compare accumulation rates between sites, normalization of the SAR’s is

achieved using focusing factors in the equation:

FCSARij (pg/m2 yr) = SARij /1=Fj (10)

where FCSARij equals the focusing corrected sediment accumulation rate

for the ith metal at the jth site and FF,- equals the focusing factor for the jth

29

site.

SAR’s and FCSAR’s are shown in Figure 5. Lake Ontario F CSAR’s
are the highest followed by lakes Michigan and Superior. Average
F CSAR’s for lakes Ontario, Michigan and Superior are .08, .03 and .006
g/mzyr, respectively, and are slightly lower than previous estimates which is
expected considering loading has been currently decreasing (Christensen
and Osuna, 1989). To assess the effects of sediment focusing and the
resulting reductions in variability between sites, CV’s of the SAR’s and the
F CSAR’s were calculated for each lake. CV’s for uncorrected and
corrected accumulation rates show that the only significant reductions in
variability which occurred using focusing factors were in Lake Superior.
This reduction implies that the lake may be recently well-mixed with
respect to zinc, dominated by atmospheric deposition or both. Likewise, it
can be suggested that relatively similar coefficients of variation of SAR’s
and F CSAR’s in lakes Michigan and Ontario show that sediment focusing is
not the dominant process responsible for intralake variations in SAR’s.
Therefore, loadings to lakes Michigan and Ontario recently are either
dominated by multiple localized sources, not well-mixed or a combination

of both. Another way of examining variations in sediment accumulation

30

.0030 0200380000 0:08:00 0£N .m 0.5me

$003 .3 00 0003m 80m 000.— :0060000 05:30:00 H II
aoumtg 00 00205000 0000 5:038:80 E08600 000000000 36008 n >0m<mom
00:30? 00 000005000 000.— couflafisoom 0008600 n >0m<m

0000M 003038000< 000000000 § 003% couflsfizooxw 00000000003 _ _
.3‘ v .m .m A V .2 .m‘ v.3
«8 0mm <33 200 :20 22.530 03 M00 20» 83 82302 00.8 0.5

 

 

 

 

 

 

 

 

 

 

 

 

(M/zur/fi) 911m uonemumoov ruourrpos

Ni-Sfi

A
. ‘
K .

 

 

  

\

LO
Q

 

 

 

 

O
‘—

 

 

 

o
“l

I

3% u >32QO
:00 n 505
02300

 

 

$3 u >0m<m0m
°\com n 55%
€0.32:

:80 I 55.000
:2 n 50%
030095

In
C‘!

llllllllllllll llll|llll||llll

llllllIllllllqlllllllll'lll
llllllllllllll lllllllll
llllllllllllllllllllllllllll

llllllllllllllllllllllllllll
IO
‘-

llllllllllllllllllllllllllll

 

 

 

 

o
9'?

 

31

rates is comparison of FCSAR’s between each site. As with contaminant
inventories, the uncertainty of each FCSAR can be determined through error
propagation of the parameters used in the calculation. Uncertainty of each
FCSAR was calculated from the equation (Kolak et al., in press; modified

from Bevington, 1969):

2
§3FCSAR - = §3§A_R + S_E " 2—*SSARFF (1 1 )
FCSARZ SAR2 FF2 SAR*FF

where 0' is equal to the standard deviation of the indicated parameter. The
last term on the right hand side of equation (1 1) represents the covariance of
SAR with F F. As with zinc inventories, linear regression yields no
correlation (r2 = .23) between SAR’s and FF ’s which reduces the third term

of the equation to zero. Equation (11) can be rewritten to isolate the

standard deviation of the FCSAR:

sFCSAR =+/-(FCSAR2 * (SSARZ/SARZ + OPE/1:132» V2 (12)

The uncertainties associated with each FCSAR were calculated using

32

equation (12) and are plotted in Figure 5. In Lake Superior, examination of
uncertainties indicates that intralake variation is not significant suggesting
recent dominance of a single source such as the atmosphere, well developed
mixing or both. Site 1383 shows no recent zinc accumulation but may be
suspect due to possible sampling error because unlike other stations, this
station was sampled using a gravity-coring device. Gravity cores can cause
disturbance to the surface sediments because of (l) preceding hydraulic
shockwaves in front of the core as it is lowered or (2) tilting of the core tube
as it is enters the sediment (Blomqvist, 1985). The introduced sampling
error can affect the 21"Pb profiles and therefore, the mass sedimentation rates
used in the calculation of FCSAR’s.

Variations in uncertainties associated with lakes Ontario and
Michigan, unlike Lake Superior, appear to be highly significant. In Lake
Michigan, the majority of variability can be attributed to sites 70M and
LMl 1. The F CSAR at site LMll is suspect in that 2"’Pb profiles and the
resulting dates may indicate physical disturbance of the core due to either
erosion or slumping. The variation caused by site 70M suggests that Lake
Michigan is recently not well-mixed with respect to zinc. The lack of

mixing may result from the presence of bathymetric barriers which isolate

33

site 70M from the remainder of the lake (Simcik et al., 1996). Intralake
variability of FCSAR’s in Lake Ontario, primarily due to site LO40A, also
may be attributed to the recent presence of multiple sources, the absence of
lake-wide mixing or both. The suggestion of multiple sources is interesting
considering studies that found the dominant sources of heavy metal and
organic contaminants to Lake Ontario were the Niagara River and Hamilton
Harbour (Eisenreich et al., 1989; Thomas, 1983; Nriagu et al., 1983). The
variation of site LO4OA may be caused by the relatively large distances of
this site from riverine discharges and exclusion from lake-wide circulation
patterns (Pickett and Bermick, 1977). Interlake differences of F CSAR’s
indicate that the Great Lakes region is not presently dominated by a single
uniform source of zinc. Atmospheric deposition may still play a significant
role but may be in the form of a concentration gradient as indicated by three

distinct F CSAR ranges, each associated with a particular lake.

Historical Accumulation Rates of Zinc
Historical trends in F CSAR’s for all three lakes show increases in
zinc accumulation rates from approximately 1920 to 1970 (Figure 6). Lakes

Michigan and Ontario have decreases in FCSAR’s from 1970 to the present

34

Year

2000
1990
1980
1970
1960
1950
1940
1930
1920
1910
1900
1890
1880
1870
1860

Figure 6. Temporal variations of zinc accumulation rates

 

 

 

 

 

: I I I l I I I I I I I T l I I :
Z— 4. —I
g- a 2-
- b _
E“ 3*
i ”t j
;_ 0 L040 A FCSAR _:
E I LM68K FCSAR E
:— -NOAA3 FCSAR '3
C 1 l J l l l 1 LI 1 l l l l l I l l l l l l l :
0.0 0.2 0.4 0.6 0.8 1.0
Normalized FCSAR

from lakes Ontario (LO4OA), Michigan (LM68K) and
Superior (N OOA3).

35

1.2

while Lake Superior has relatively constant F CSAR’s during the same
period. Zinc accumulation rates increase and decrease the fastest in Lake
Ontario and the slowest in Lake Superior which is expected considering
Lake Ontario is most affected by anthropogenic inputs and Lake Superior
the least affected. Sites LM18, LM19, LS1383 and DTL have constant zinc
concentrations at the tops of the cores suggesting that loading has not
changed, however, these sites may have large mixing zones resulting in
profile homogenization (Eisenreich et al., 1989; Robbins et al., 1975).
Inspection of the temporal onset of anthropogenic loading in Lake
Michigan shows the southernmost sites, LM18 and LMl 1, have early onsets
during the 1850’s, the mid-latitude sites, LM47s and LM68k, have later
onsets around the 1920’s and site 70m, the most northern site in Lake
Michigan, has the earliest onset around 1815. These variations imply that
either loading was initially dominated by multiple localized sources, the
lake is not well-mixed with respect to zinc or both. Concentration profiles
and onsets of zinc pollution in the lake agree with the work of Christensen
and Goetz (1987) which had onsets starting around 1894 in the southern
part of the lake and 15 to 30 years later in the northern and central parts.

Similar studies of Lake Michigan using the same sediment cores have also

36

found the same patterns with Cu (Kolak et al., in press), Pb, Cd and Hg

(Long et al., 1995).

Atmospheric Deposition of Zinc

Similar interlake FCSAR’s and large reductions in CV’s suggest a
recent single uniformly loading source of zinc to the lakes. This source of
the loading is most likely the atmosphere due to the lack of any other source
of such a large spatial scale. Atmospheric deposition to the Great lakes is
enhanced by large lake surface areas within relatively small drainage basins
resulting in a large fraction of the hydrologic input to be from direct
precipitation. Furthermore, some processes that release metals such as Pb
and Cd into the atmosphere (e. g. combustion, smelting) also release zinc
(Gatz et al., 1989; F ingleton and Robbins, 1980; Winchester and Nifong,
1971). Trace metals enter the atmosphere from sources such as wind blown
dust, fossil fuel combustion, metal ore processing and refuse incineration
(Dutkiewicz et al., 1987). The Great Lakes are dominated by easterly
heading maritime and continental polar winds which carry metals from
large urban/industrial sources over the lakes, such as Chicago and Toronto

loadings to Lake Michigan and Ontario, respectively (Environment Canada,

37

1987). Typical residence times of zinc aerosol in the atmosphere range
from 4 to 7 days making long distance transport possible (Norton et al.,
1981). Over the lakes the contaminants are transferred to sediments through
wet and dry deposition of atmospheric particulate matter (Eisenreich, 1980).
The areas of heaviest atmospheric deposition of trace metals in the Great
Lakes are located in areas in close proximity to major urban/industrial
centers such as southern Lake Michigan (close to Chicago and Gary) and
Lake Ontario (near Toronto, Hamilton and Rochester) (Sweet et al., 1998;
Simcik et al., 1996; Davis and Galloway, 1981; Winchester and Nifong,
1971)

Intralake and interlake variations in zinc inventories and
accumulation rates are used to qualitatively examine the role of atmospheric
deposition in contaminant input. To quantitatively determine the recent role
of attnospheric deposition it is necessary to compare zinc FCSAR’s with
current zinc atmospheric deposition rates (Eisenreich, 1980). Atmospheric
deposition rates, including both wet and dry deposition, were obtained from
the Integrated Atmospheric Deposition Network (IADN) (Figure 1) from
1993 to 1994 because this network utilized current sampling protocols and

had a regional distribution of sampling stations. Estimated atmospheric

38

deposition rates at IADN sites for Lake Superior and Michigan are taken
from Sweet et al. (1998) and data for Lake Ontario is from Hoff (pers.
comm).

Table 1 shows the comparison of F CSAR’s with atmospheric
deposition rates. The atmospheric contribution to the total metal input is
calculated by dividing current estimates of atmospheric deposition rates for
the lake by the FCSAR at each site. The percentages of atmospheric
contribution obtained were Lake Superior (100%), Lake Michigan (IO-61%)
and Lake Ontario (14-50%). The values for Lake Superior agree well with a
previous study which found atmospheric fallout from numerous mining and
smelting operations to be a dominant source of zinc (Nriagu et al., 1996).
The results also agree with other studies of Lake Superior which found
input of Pb, Hg, Cd and PCB to be dominated by the atmosphere (Long et
al., 1995; Eisenreich et al., 1992). Site DTL has a lower atmospheric
contribution of 83% which may be caused by input fi'om the Duluth River
compared to other intralake sites.

The large variability in estimates for Lake Michigan may be
attributed to a number of factors. The dominance of atmospheric fallout

from Chicago and the surrounding industrial complex as a loading source in

39

Table 1. Percent atmospheric contribution of zinc relative to total inputs.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. Atrn. De . A h ri
3353:; (Sign Rage 1’ £3328in:

(lug/m lyr) (%)

DTL 8497 8 800A 100
SJE - 2 10549 8800A 83

LS 13 91 6704 8800A 100

NOAA3 615 1 8800A 100
LM70M 9801 6000A 61
LM68K 50460 6000A 1 2
LM47S 59394 6000A 10
LM 1 8 37436 6000A 1 6
LM 1 9 42 1 74 6000A l 5

LM68K 50460 50000“ 100
LM47S 59394 45000“ 76

LMl 8 3 7436 9000* 100
LM19 42174 21000* 50
L019 79872 15737A 20
LO40A 30918 15737A 50
E30 104470 15737A 15
G32 111863 15737A l4

 

 

" atmospheric deposition values obtained from Sweet et al. (1998) and
Hoff (personal communication)

* atmospheric deposition values obtained from Christensen and Osuna
(1989)

40

the southern basin has been well documented (Simcik et al., 1996; Gatz et
al., 1989; F ingleton and Robbins, 1980; Winchester and Nifong, 1971).
Eisenreich (1980) estimated a percent atmospheric contribution of 67% for
zinc loading to Lake Michigan. The low contributions found in this current
study may be from the use of a single atmospheric deposition rate for the
entire lake even though it is known to be significantly enhanced in the south
from urban and industrial activities (Cole, 1990; F ingleton and Robbins,
1980). The underestimation of the calculated percentages and the
associated error of using one atmospheric depositional value is supported by
the relatively large atmospheric contribution of site 70M and it’s proximal
location to the IADN sampling site for Lake Michigan (Figure 1). Using
recent atmospheric deposition rates from the study of Christensen and
Osuna (1989) for Lake Michigan gives expectedly larger atmospheric
contributions ranging fiom 50 to 100%.

As with Lake Michigan, the variability in Lake Ontario may result
fiom the use of a single atmospheric deposition rate for the entire lake even
though atmospheric fallout is most likely enhanced in areas adjacent to
Toronto and Hamilton. If atmospheric deposition rates are correct,

however, the relatively low atmospheric contributions of sites L019,

41

G32 and E30 agree well with other studies which found contaminant input
in Lake Ontario to be dominated by effluents from the numerous chemical
industries on Niagara River (Eisenreich et al., 1992; Thomas, 1983; Pickett
and Bermick, 1977) and iron and steel plants at Hamilton Harbour (Nriagu
et al., 1996). The large distance of site LO4OA from the mouth of the
Niagara River and it’s relative exclusion from effluent circulation patterns
(Thomas, 1983) may be a possible explanation for the large atmospheric
contribution compared to other intralake sites. The need for more accurate
estimates of atmospheric deposition rates and errors associated with
previous estimates has been well noted (Sweet et al., 1998; Hoff, 1994;
Arimoto, 1989). Better measurements of atmospheric deposition rates will
allow more accurate quantification of atmospheric contributions to
sediments.

Inventory and FCSAR calculations showing three distinct ranges,
each associated with a particular lake, increasing in the order of Lake
Superior, Lake Michigan and Lake Ontario, as discussed previously may
suggest the dominance of the atmosphere in the form of a depositional
gradient. The presence of a regional depositional gradient, however, is

unlikely when considering the sources of zinc to the lakes, the possible

42

orientation of the gradient, and regional wind patterns. If an atmospheric
depositional gradient was responsible for the interlake trends of inventories
and FCSAR’s, the primary source of zinc to the Great Lakes region must be
adjacent to Lake Ontario (which has the highest zinc loading), with Lake
Michigan and then Superior being located down gradient. Eastward moving
air masses dominate the region and generally cause gradients to decrease
towards the east and not westward as indicated by inventories and

F CSAR’s. Differences between the orientation of the regional wind
direction and interlake variations show an atmospheric depositional gradient
could not be responsible for observed interlake trends of inventories and

F CSAR’s.

Possible Sources of Zinc

Historical input functions of contaminants may be identified through
temporal similarities of F CSAR’s with specific production and consumption
data. Edgington and Robbins (1976) found sedimentary profiles of Pb
similar to production and consumption data for gasoline. The comparison
of accumulation profiles with production/consumption data has also been

used to evaluate potential sources of PCB, DDT, HCB and Mirex

43

(Eisenreich, 1989), PAI-I’s (Simcik et al., 1996), Pb, Hg, Cd and As (Long
etal,l995)

Numerous anthropogenic activities, including fossil fuel combustion
and industrial manufacturing, contribute to zinc loading in the Great Lakes.
The focus of a study by the EPA (1987) to identify the major source of zinc
emissions found potentially large sources in the manufacture of primary
zinc, secondary zinc, brass, iron and steel. The primary centers of these
activities are the large. urban and industrial centers adjacent to the lakes such
as Chicago and Gary around Lake Michigan, and Toronto and Hamilton
around Lake Ontario. Fingleton and Robbins (1980) attributed high
concentrations of zinc in winds over Lake Michigan (which originated from
Gary, Indiana), to the many metal processing industries in that area.
Winchester and Nifong (1971) attributed large inputs of zinc to the southern
lake basin to the iron and steel manufacturing industries located adjacent to
the lake. Previous studies of zinc sources to Lake Ontario indicate effluents
from iron and steel mills at Hamilton and Nanticoke (Environment Canada,
1987; Nriagu et al., 1983), effluents fi'om chemical industries on the
Niagara River (Mudroch, 1983) and atmospheric fallout from Toronto

(Nriagu et al., 1983). Another possibly significant source of zinc is the

44

deposition of fly ash originating from coal fired power plants (Pacyna,
1986; Kaakinen et al., 1975) which have been historically dominant in
energy generation for the Great Lakes region (Cirillo, 1975).

Temporal trends in production and consumption of zinc were
examined through the compilation of data obtained fi'om the Minerals
Yearbook and the Annual Energy Review. Data examined for both the
United States and Canada are listed in Table 2. To directly compare
temporal trends in FCSAR’s with production and consumption histories,
both sets of data were normalized to their maximum values respectively,
and plotted as a function of time (Figure 7). Sites NOAA3, LM68K and
LO40A were chosen as lake representative sites for comparison based on
the integrity of their 21"Pb profiles. It is important to note that when
comparing FCSAR’s to production/consumption histories, general trends
must be examined because changes in production/consumption which occur
in time periods shorter than the intrinsic resolutions of the cores will not be
reflected in FCSAR’s. In Lake Superior, historical trends in FCSAR’s
closely resemble trends in zinc smelter production for the United States and
less closely for trends in consumption of zinc used for diecasting. The

resemblance of these trends are not surprising considering the close location

45

Table 2. Productive and consumptive uses of zinc as possible
source functions.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

46

For the United States:
Bmducttqn 0000000911
Bituminous Coal Bituminous Coal
Anthracite Coal Anthracite Coal
Total Coal Total Coal
Gasoline Gasoline
Zinc Ore Zinc Ore
Slab Zinc Slab Zinc
Secondary Zinc Secondary Zinc
Redistilled Zinc Zinc in Brass
Distilled/Electrolytic Zinc in Diecasting
Zrnc
Rolled Zinc Zinc in Pigments
Zinc Dust
Smelter Production
Pig Iron
Steel
Chemicals
For Canada:
Imagination
Total Zinc
Pig Iron
Steel

 

.88 coumfismGOQEouoswoa o0 “000.000.0000

000.01. 00000 06 00. 8:0 3.000000000000000 00 00000 00000300030000 0.00.0068 ofiN .N. 000%.”.

$03. nom0oswo0mEou£=E=00< 03:08.8 Z
0.0 3 03 we to No od

 

14.0.410_0.Jl_0_________.
0280 <83 0
ms0§ofil
0.200.605 ofiN .80... .'

llIl'lIllll'll'll[VIIIIIIVVIIVIII'Y'

VlllIIIVIIIVVVIV|VI‘TIIYIIYVIIlllvlll

 

p_———_>—___—_______———_

  

1111111111111

111111111I111111111l1111l1111l111

    

 

 

2000.00 00.03

Omwfi
cow.
chm.
oww.
cam.
com.
ofim.
omafi
oma.
ova.
owo.
com.
Ohm.
owa.
cam.
ooom

N.. o.. wgvao «xo NAVAfio
.omw.

 

1. .. _. . ._ .. ._ .. ._ .. ._ ._ .

m<m0m Mme—>5 .
. 000000.605 60% I
1 002003005 :00. mil

 

vL—~——-—--———~—~_——_~_

.82
not:

oww.
oaw.
com.

.22
”82

omo.

“ova.
Memo.

coo.
ova.

Howa.
Hcam.

 

Hooom

 

$000022 8.3

 

'1

 

1.. ._. ..—. .444 #4_ ..._ ...

M<mom m<<OZ .
waumaooal
.0800:on 000.00% I

 

4
L
L

l
I

l

 

 

0200msm 00.04

N. o; wgvcgvvgvmgvogv

oww.

82:
H2.:
6%.
.80:
W82
65
Mona
5.80
630
meme
W82
W22
W230
W800
.ooom

1..____._._____._..._._..

199A

47

of both zinc smelters and metal processing plants to Lake Superior. Both
F CSAR’s and smelter production exhibit increases approximately from
1925 to 1970. After 1970, however, FCSAR’s are constant to the present
even though total zinc production decreases during the same period. The
discrepancy between the two trends may have multiple explanations.
Mixing may be responsible for the constant trends in the FCSAR’s but
would not be reflected in the production history. Another explanation is
total production controlled zinc input to Lake Superior from 1925 to 1970
and after this time another control began to dominate (e. g. gasoline
combustion). Diecasting, another use of zinc which follows the increase in
FCSAR’s from 1925 to 1970, is an interesting source considering it was the
principal use of zinc in the United States from the 1950’s to the 1970’s
accounting for 39% of total zinc consumption (Dammert and Chabra, 1990).
Lake Michigan F CSAR’s closely match production data for both steel
and pig iron in the United States. The match of both pig iron and steel is
expected because pig iron is the raw material used in producing steel. Both
FC SAR’s and production data show rapid increases from approximately
1925 to 1975, decreases from 1975 to 1985, and no changes from 1985 to

the present. The agreement of Lake Michigan FCSAR’s with data from the

48

iron and steel industry is not surprising considering many studies which
found the origin of zinc input to be from atmospheric fallout of emissions
from steel producing industries located on the lakes’ southern perimeter
(U SEPA, 1987; Eisenreich, 1980; Winchester and Nifong, 1971).
Furthermore, the high percentages of atmospheric contribution (previous
section) to Lake Michigan also suggest the importance of an atmospheric
source.

Accumulation trends in Lake Ontario F CSAR’s resemble trends in
United States for total zinc production and zinc consumed in diecasting.
These sources are, however, unlikely as controlling inputs because zinc
input to Lake Ontario is known to be dominated by effluents from Niagara
River chemical industries and from Canadian iron and steel plants located in
Harniton Harbour. Data from these sources were not available at the time of
this study thereby preventing direct comparison. The low percentage of
atmospheric contribution to Lake Ontario also supports the dominance of
effluent inputs to the lake. Production figures of zinc, pig iron, and steel in
Canada also do not match zinc FCSAR profiles. The lack of correlation of
Lake Ontario profiles with any single production/consumption trend from

either the United States or Canada is expected considering the multiple

49

sources to this lake.

Temporal trends in FCSAR’s from all three lakes may also be a
function of enactment of environmental regulations regarding pollutant
emissions during the late 1960’s and early 1970’s. Similar loading trends,
when observed in peat bogs, have been attributed to increased industrial
emissions and later limiting regulations (Cole et al., 1990; Rapaport and
Eisenreich, 1988; Schell, 1986). Emissions in the Great Lakes region were
regulated by the passage of the Clean Air Act in the United States in 1963
and amended in 1970 and 1977, and a similar law enacted during the same
period in Canada (Boubel et al., 1994). The steel industry, as a possible
source function, experienced a shifi fi'om the use of open hearth furnaces to
basic oxygen furnaces during the 1960’s which reduced particulate
emissions already being lowered through the introduction of baghouses and
scrubbers in production plants (Boubel et al., 1994). Sites not dominated by
atmospheric deposition, but by input fiom industrial effluents also
experienced a reduction in input due to the passage of the Clean Water Act
in 1972 (Patrick et al., 1992). Both passing of environmental regulations
and temporal trends in production may be responsible for the decrease in

zinc F CSAR’s since the late 1960’s and early 1970’s. The relative

50

contribution of each of these inputs to FCSAR profiles, however, is
unknown thereby preventing attribution of a sole source of zinc to the Great
Lakes region.

The comparison of FCSAR’s with production/consumption data has
many potential limitations. Christensen and Osuna (1989) noted: 1) the
possibility that metal use in the Great Lakes is not a constant fraction of the
US. consumption 2) changes in metal use may cause changes in emission
rates and 3) advances in the removal of particulates by polluting industries
may decrease atmospheric loading of the metals. F6rstner and Wittman
(1979) noted that multiple uses of heavy metals leads to difficulties in
assigning water pollution to one source conclusively. The same study also
found many minor uses of zinc which may play an important role in
contamination of the Great Lakes such as use in paper and cement
production, urban stormwater runoff, release associated with refuse

incineration and the wearing of automobile tires.

Comparison With Other Contaminants in the Great Lakes
Many studies of trace metal inputs to 1) Lake Superior found

atmospheric dominance for Zn, Cd and Hg inputs (this study; Long et al.,

51

1995) 2) Lake Michigan found dominance of atmospheric fallout from
Chicago and Gary for Zn and Cd (Gatz et al., 1989; Eisenreich, 1980) 3)
Lake Ontario found effluent dominance of the Niagara River and Hamilton
Harbour for Zn, Cd and Hg (Nriagu et al., 1996; Mudroch, 1983). The
similar sources of these trace metals to all three lakes, respectively, should
also be evident in this study through similar accumulation characteristics
and may also suggest similar environmental controls (i.e. regulations).
These accumulation characteristics (Table 3) include interlake trends in
both metal inventories and accumulation rates, temporal trends in
accumulation rates, percents of atmospheric contribution and sources
suggested through comparison with production/consumption histories. Zinc
was compared to contaminants including As (Long et al., 1995), Cr (J eong,
1994), Cu (Kolak et al., in press), and organic contaminants such as PCB,
DDT and PAH’s (Golden et al., 1993; Simcik et al., 1996).

Similar comparisons arose particularly with Cd and Hg (Table 3).
Similar transport of zinc and cadmium via aerosols is expected because of
their close geochemical association (Christensen and Chien, 1981; Nriagu,
1980) such as their common occurrence and release in the smelting of

sulfide ore deposits (F 6rstner and Wittman, 1979) and high affinity for

52

Table 3. Comparison of zinc accumulation characteristics

 

 

 

 

 

 

 

 

 

 

 

 

 

 

with those of cadmium and mercury.
1m
lat-Lara. m 1111111111.: m
lnventgy Accumulation FCSAR Amber—i1: Sources
1110 Bic; m m
110. rs WW
&%0 15 = 100% steel
= 0 .
LO>LM>IS 1975; all ID>LM>LS 1113: 330? W
decrease to the ° poducuon/
present steel
[O m
urtil 1966;
IM increases IS = 100%
I10>IM>LS until 1970; LO>LM>LS [M = 47% snelt'mg
both decrease LO =24%
to the preset
LO, LS
irn‘eases urtil
1966; LM LS = 100% 1 . l
LO>>LM>LS increases urtil ID>>LM>LS LM = 100% .
1964; all 10 = 14% poductron
decrwe to the
11661
IS = Lake Superior
[M = Lake Michigan
L0 = lake Ontario

53

organic matter (Nriagu, 1980). Schmidt and Andren (1984) compiled the
results of trace metal loading studies in the Great Lakes and found the most
similar patterns with zinc and cadmium. In this study, accumulation
characteristics common to both zinc and cadmium include decreasing
inventories and accumulation rates in the order of Lake Ontario, Michigan
and Superior, similar peak dates in Lake Ontario, and similar percent
atmospheric contributions in all three lakes. The similar environmental
behavior of zinc and cadmium is likely also because both share common
sources which are the metals manufacturing and smelting industries in Lake
Michigan (Forstner and Wittman, 1979; Winchester and Nifong, 1971) and
the chemical and metal manufacturing industries in Lake Ontario (Mudroch,
1983). Zinc and mercury also show very similar accumulation
characteristics in the Great Lakes region which is unexpected because of
their different sources. Interlake trends in both inventories and
accumulation rates of zinc and mercury show decreases in the order of lakes
Ontario, Michigan and Superior, respectively.

One difference between the two metals is that mercury inventories
and accumulation rates in Lake Ontario are significantly larger than those of

Lake Michigan compared to zinc inventories and accumulation rates which

54

are only slightly larger in Lake Ontario than in Lake Michigan. This
discrepancy may be attributable to the different sources of mercury and zinc
to Lake Ontario. The chloro-alkali plants present on the Niagara River
dominate mercury inputs (Long et al., 1995) and zinc inputs are contributed
by both metal processing industries in Hamilton and chemical industries on
the Niagara River (this study). Similar characteristics of zinc and mercury
are also large atmospheric loadings in lakes Superior and Michigan and

smaller atmospheric loadings in Lake Ontario.

55

CONCLUSIONS

The objective of this study was to perform a regional assessment of
zinc accumulation in the Great Lakes and provides the following
conclusions (Table 4): (1) anthropogenic inventories and sediment burdens
are highest in Lake Ontario followed by Lake Michigan and Lake Superior;
(2) intralake inventory comparisons indicate Lake Ontario is historically
well-mixed with respect to zinc and lakes Superior and Michigan are not;
(3) temporal trends in zinc accumulation rates show that loading is currently
decreasing in lakes Michigan and Ontario and is constant in Lake Superior;
4) intralake comparisons of accumulation rates indicate that recently Lake
Superior is well-mixed and lakes Michigan and Ontario are not well-mixed
with respect to zinc; (5) atmospheric deposition accounts for 100%, 64%,
and 20% of zinc input to lakes Superior, Michigan, and Ontario,
respectively; (6) sources of zinc to Lake Superior are a function of zinc
produced by smelting or consumed in diecasting, sources to Lake Michigan
originate from the iron and steel industries in the Chicago and Gary area,
and sources to Lake Ontario are effluents from chemical industries on

Niagara River and iron and steel industries at Hamilton Harbour; (7)

56

Table 4. Environmental summary of zinc in the Great Lakes.

 

 

 

 

 

 

 

 

 

tarmac: 5.00.9110; Michigan 01.118110
Inventory Low Higher Highest
Historically
Well-Mixed? No No Yes
Recently
Well-Mixed? Yes N° N°
% Atmospheric o o 0
Contribution 100 /° 64 /° 20 /°
Recent
Accumulation Constant Decreasing Decreasing
Trends
Smel tin Iron and Chemical
Possible Sources Diecastirgr, Steel and Steel
g Production production
Similar
Accumulation Cd, Hg Cd, Hg Cd, Hg
Characteristics

 

 

 

 

57

 

decreasing accumulation rates of zinc in the Great Lakes may be caused by
environmental regulations and/or trends in production and consumption but
the contribution of each requires further study; (8) comparison of zinc
accumulation characteristics with those of other metals suggests similar

environmental controls with respect to cadmium and mercury.

58

REFERENCES

Andrew R.W. (1970) Distribution of taconite tailings in the sediments of
the western basin of lake Superior. Investigations by the Stafi” of the
National Water Quality Laboratory. April, 1970

Arimoto R. (1989) Atmospheric deposition of chemical contaminants to
the Great Lakes. J. Great Lakes Res. 15(2), 339-3 56.

Atlas R.M., and Bartha R. (1993) The oxygen cycle. In Microbial
Ecology: Fundamentals and Applications, Third Edition. pp.309-
310. The Benjamins/Cummings Publishing Company Inc., Redwood
City, California.

Bemer R.A. (1980) Early Diagenesis, Princeton University Press,
Princeton, NJ.

Bevington RR. (1969) Data reduction and error analysis for the physical
sciences. McGraw Hill Inc., New York.

Blomqvist, S. (1985) Reliability of core sampling of soft bottom sediment
- an in situ study. Sedimentology 32, 605-612.

Boubel R.W., Fox D.L., Turner DB, and Stern AC. (1994) Sources of air
pollution. In Fundamentals of Air Pollution. pp.72-96. Academic
Press, San Diego.

Cahill R.A., and Shimp NP. (1984) Inorganic contaminants in Lake
Michigan sediments. In Toxic Contaminants in the Great Lakes. (ed
Nriagu J .O., and MS Simmons) pp.393-423, John Wiley & Sons,
NewYork.

Chan W.H., and Lusis MA. (1986) Smelting operations and trace metals
in air and precipitation in the Sudbury Basin. In Toxic Metals in the
Atmosphere (ed J .O. Nriagu, and CI. Davidson) pp.113-143, John
Wiley & Sons Inc., New York.

59

Christensen ER. and Chien N. (1981) Fluxes of arsenic, lead, zinc and
cadmium to Green Bay and Lake Michigan sediments. Environ. Sci.
T echnol. 15, 553-58.

Christensen E.R., and Goetz RH. (1987) Historical fluxes of particle-
bound pollutants from deconvolved sedimentary records. Environ.
Sci. T echnol. 21, 1088-1096.

Christensen ER. and Osuna J .L. (1989) Atmospheric fluxes of lead, zinc,
and cadmium from frequency domain deconvolution of sedimentary
records. J. Geophys. Res. 94(C10), 14585-14597.

Cirillo RR. (1975) Fossil-fueled—fired power plant impacts on the Great
Lakes. In Proc. Second Federal Conference on the Great Lakes,
pp.77-8 1.

Dammert, A.J. and Chabra J .G.S. (1990) The lead and zinc industries, long
term prospects. The World Bank, Washington DC.

Davis AD. and Galloway J .N. (1981) Atmospheric lead and zinc
concentrations into lakes of the eastern United States. In Atmospheric
Pollutants in Natural Waters (ed. S.J. Eisenreich) pp.401-407, Ann
Arbor Science Publishers Inc., Ann Arbor, Michigan.

Dutkiewicz V.A., Parekh PP, and Husain L. (1987) An evaluation of
regional elemental signatures relevant to the northeastern United

States. Atmospheric Environment 21(5), 1033- l 044.

Edgington D.N., and Robbins LA. (1976) Records of lead deposition in
Lake Michigan sediments since 1800. Environ. Sci. Technol. 10(3),
266-274.

Eisenreich S.J., Capel P.D., Robbins J .A., and Bourbonneire R. (1989)
Accumulation and diagenesis of chlorinated hydrocarbons in

lacustrine sediments. Environ. Sci. T echnol. 23(9), 1116-1126.

Eisenreich S.J., Metzer N.A., and Urban NR. (1986) Response of
atmospheric lead to decreased use of lead in gasoline. Environ. Sci.

60

Technol. 20(2), 171-174.

Eisenreich S.J., Baker J .E., Franz T., Swanson M., Rapaport R.A., Strachan
R.A., and I-Iites R.A. (1992) Atmospheric Deposition of
Hydrophobic Organic Contaminants to the Laurentian Great Lakes.
In Fate of Pesticides and Chemicals in the Environment, pp. 51-78,
John Wiley & Sons Inc., New York.

Eisenreich SJ. (1982) Atmospheric role in trace metal exchange at the air-
water interface. J. Great Lakes Res. 8(2), 243-256.

Eisenreich SJ. (1980) Atmospheric input of trace metals to Lake
Michigan. Water, Air, and Soil Pollution 13, 287-301.

Environment Canada (1987) The Great Lakes, An Environmental Atlas and
Resource Book Joint publication by Environment Canada, USEPA,
Brock University and Northwestern University.

F ingleton DJ, and Robbins J .A. (1980) Trace elements in air over Lake
Michigan near Chicago during September, 1973. J. Great Lakes Res.

6(1), 22-37.

Fbrstner U., and Wittmann G.T.W. (1979) Metal Pollution in the Aquatic
Environment. Published by Springer-Verlag, New York.

Gatz D.F., Bowersox V.C. and Su J. (1989) Lead and cadmium loadings to
the Great Lakes from precipitation. J. Great Lakes Res. 15(2), 246-
264.

Glaser J .A., Foerst D.L., McKee G.D., Quave S.A., and Budde W.L. (1981)
Trace analyses for wastewaters. Enviro. Sci. and Tech. (15)]2, 1426-
1434.

Golden K.A., Wong C.S., Jeremaison J .D., Eisenreich S.J., Sanders G.,
Hallgren J ., Swackhammer D.J., Engstrom DR, and Long D.T.
(1993) Accumulation and Preliminary Inventory of Organochlorines
in Great Lakes Sediments. In Proceedings of the First international
specialized conference on contaminated aquatic sediments, historical

61

records, environmental impact and remediation. (eds. E.R.
Christensen, D.N. Edgington, and J .P. Giesy). Water Science and
Technology 28(8-9), 19-31.

Hart BA. (1982) Uptake of trace metals by sediments and suspended
particulates: a review. Hydrobiologia (91), 299-313.

Hewitt A.D., and Reynolds CM. (1990) Dissolution of metals from soils

and sediments with a microwave-nitric digestion technique. Atomic
Spectroscopy 1 1, 187-192.

Hilton J., Lishman JP, and Allen P.V. (1986) The dominant processes of
sediment distribution and focusing in a small, eutrophic, monomictic
lake. Limnol. Oceanogr. 31(1), 125-133.

Hoff R.M. (1994) An error budget for the determination of the atmospheric
mass loading of toxic chemicals in the Great Lakes. J. Great Lakes
Res. 20(1), 229-239.

International Joint Commission (1978) Environmental management
strategy for the Great Lakes system. Final report to the International
Joint Commission fi'om the International Reference Group on Great

Lakes Pollution from Land Use Activities (PLUARG), July 1978.
Windsor, Ontario.

J eong S. (1994) Spatial and temporal variations of chromium in sediments
of the Great Lakes. M.S.Thesis, Michigan State University.

Johnson MG. and Nicholls K.H. (1988) Temporal and spatial trends in
metal loads to sediments of Lake Simcoe, Ontario. Water, Air and
Soil Pollution 39, 337-354.

Kaakinen J .W., Jorden R.M., Lawasani M.H., and West RE. (1975) Trace
element behavior in a coal-fired power plant. Environ. Sci. T echnol.

9(9), 862-869.

Kelly T.J., Czuczwa J .M., Sticksel PR, and Sverdrup GM. (1991)
Atmospheric and tributary inputs of toxic substances to Lake Erie

62

J. Great Lakes Res. 17(4), 504-516.

Kemp A.L.W., Anderson T.W., Thomas R.L., and Mudrochova A. (1974)
Sedimentation rates and recent sediment history of Lakes Ontario,

Erie, and Huron. J. Sedimentary Petrology 44, 207-218.

Kemp A.L.W., Williams D.H., Thomas R.L., and Gregory M.L. (l978a)
The chemical composition of the sediments of Lakes Superior and
Huron. Water, Air and Soil Pollution 10, 381-402.

Kemp A.L.W., Dell CL, and Harper N.S. (1978b) Sedimentation rates and
a sediment budget for Lake Superior. J. Great Lakes Res. 4(3-4),
276-287.

Kesler-Arnold K.A. (1990) Background concentrations of metals and
cyanide in lower Michigan soils. 44‘” Purdue Industrial Waste
Conference Proceedings. Published by Lewis Publishers Inc.,
Chelsea, Michigan.

Kolak J .J ., Long D.T., and Beals T.M. Anthropogenic inventories and
historical and present accumulation rates of copper in Great Lakes
sediments. in press to Applied Geochemistry.

Long D.T., Sitarz W., Eisenreich SJ. and Swackhammer D.L. (1995)
Atmospheric deposition of toxic contaminants to the Great Lakes:
Assessment and Importance Volume 111. Metal concentrations in
sediments of the Great Lakes. Report to the Great Lakes Protection
Fund. 82pp.

McKee J .D., Wilson T.P., and Long D.T. (1989a) Pore water profiles and
early diagenesis of Mn, Cu, and Pb in sediments fi'om large lakes. J.
Great Lakes Res. 15(1), 68-83.

McKee J .D., Wilson T.P., Long D.T., and Owen R.M. (1989b)
Geochemical partitioning of Pb, Zn, Cu, Fe, and Mn across the

sediment-water interface in large lakes. J. Great Lakes Res. 15(1),
46-58.

63

Moore J. N., F icklin W.H., and Johns C. (1988) Partitioning of arsenic and
metals in reducing sulfidic sediments. Environ. Sci. Technol. 22,
432-43 7.

Mudroch A. (1983) Distribution of major elements and metals in sediment
cores from the western basin of Lake Ontario. J. Great Lakes Res.
9(2), 125-133.

Mudroch A., Sarazin L., and Lomas T. (1988) Summary of surface and
background concentrations of selected elements in the Great Lakes
sediments. J. Great Lakes Res. 14(2), 241-251.

Norton S., Hess C.T., and Davis RB. (1981) Rates of accumulation of
heavy metals in pre and post-European sediments in New England
Lakes. In Atmospheric Pollutants in Natural Waters (ed S.J.
Eisenreich) pp.409-421, Ann Arbor Science Publishers Inc, Ann
Arbor, Michigan.

Nriagu, J .O., Lawson G., Wong H.K.T., and Cheam V. (1996) Dissolved
trace metals in lakes Superior, Erie, and Ontario. Environ. Sci.
T echnol. 30(1), 178-187.

Nriagu, J .O., Wong H.K.T., and Snodgrass W.J. (1983) Historical records
of metal pollution in sediments of Toronto and Hamilton Harbours.
J. Great lakes Res. 9(3), 365-373.

Nriagu J .O. (1980) Zinc in the Environment Part I: Ecological Cycling.
John Wiley & Sons Inc., NewYork.

Nriagu J .O. (1980) Particulate and dissolved trace Mmetals in Lake
Ontario. Water Research. 15, 91-96.

Owen R.M., and Macklin J .E. (1980) Authigenic associations between
selected rare earth elements and trace metals in lacustrine sediments.

Environ. Geol. 3, 131-137.

Pacyna J .M. (1981) Atmospheric Trace Elements from Natural and
Anthropogenic Sources. In Toxic Metals in the Atmosphere

64

(ed J .O. Nriagu, and CI. Davidson) pp.33-52, John Wiley & Sons, New
York.

Patrick R., Douglass F ., Palavage D.M., and Stewart RM. (1992) Surface
Water Quality: Have the Laws been Successful?, Princeton University
Press, Princeton, NJ.

Pickett R.L., and Bermick S. (197 7) Observed resultant circulation of Lake
Ontario. Limnol. Oceanogr. 22, 1071-1076.

Rapaport R.A., and Eisenreich SJ. (1988) Historical atmospheric inputs of
high molecular weight chlorinated hydrocarbons to eastern North
America. Environ. Sci. T echnol. 22, 931-941.

Robbins J .A. (1978) Geochemical and geophysical applications of
radioactive lead. In The Biogeochemistry of Lead in the Environment
Part A. Ecological Cycles (ed J .O. Niagru) pp.285-293,
Elsevier/North-Holland Biomedical Press, Amsterdam.

Schell W.R. (1986) Deposited atmospheric chemicals. Environ. Sci.
T echnol. 20(9), 847-853.

Shacklette H.T., and Boemgen J .G. (1984) Element concentrations in soils
and other surficial materials of the conterminous Unites States. US.
Geological Survey Professional Paper, 105pp.

Schmidt J .A., and Andren A.W. (1984) Deposition of airborne metals into
the Great Lakes: An evaluation of past and present estimates. In
Toxic Contaminants in the Great Lakes. (ed Nriagu J .O., and MS.
Simmons) pp.81-103, John Wiley & Sons, NewYork.

Simcik M.F., Eisenreich S.J., Golden K.A., Liu S.P., Lipiatou E.,
Swackhammer D.L., and Long D.T. (1996) Atmospheric loading of
polycyclic aromatic hydrocarbons to Lake Michigan as recorded in
the sediments. Environ. Sci. T echnol. 30(10), 3039-3046.

Sitarz W.J. (1996) Accumulation and inventory of mercury in sediments of
the Great Lakes. MS. Thesis. Michigan State Unversity.

65

Stigliani W.M., Doelman P., Salomons W., Schulin R., Smidt G.R.B., and
Van der Zee T.M. (1991) Chemical time bombs: predicting the
unpredictable. Environment 33(4), 4-30.

Sweet C.W., Weiss A., and Vermette SJ. (1998) Atmospheric deposition
of trace metals at three sites near the Great Lakes. submitted to
Water, Air and Soil Pollution.

Tessier A., Campbell FCC. and Bisson M. (1980) Trace metal speciation
in the Yamaska and St. Francois rivers. Can. J. Earth Sci. 17, 90-
105.

Thomas, R.L. (1983) Lake Ontario sediments as indicators of the Niagara
River as a primary source of contaminants. J. Great Lakes Res. 9(2),
118-124.

Urban N.R., Eisenreich S.J., Grigal DR, and Schurr K.T. (1990) Mobility
and diagenesis of Pb and 21"Pb in peat. Geochimica et Cosmochimica
Acta. 54, 3329-3346.

U.S.E.P.A. (1987) Zinc/Zinc Oxide preliminary source assessment. US.
Environmental Protection Agency, Office of Air and Radiation,
Office of Air Quality Planning and Standards, April 1987.

Van Loon J .C., and Beamish R.J. (1977) Heavy metal contamination by
atmospheric fallout of several F lin Flon area lakes and the relation to
fish populations. J. Fish. Res. Board Can. 34, 899-906.

Winchester J .W. and Nifong GD. (1971) Water pollution in Lake
Michigan by trace metals from pollution aerosol fallout. Water, Air
and Soil Pollut. 1, 50-64.

Wong C.S., Sanders G., Engstrom D.R., Long D.T., Swackhammer D.L.,
and Eisenreich SJ. (1995) Accumulation, Inventory and Diagenesis
of Chlorinated Hydrocarbons in Lake Ontario Sediments. Environ.
Sci. T echnol. 29, 2661-267

66

   

IIIIIIIIIIIIIIIIIIIIIIIIIIIIII

1131111111111151111111111