The influence of lake morphometry on sediment focusing

gayoldΜηχανική

21 Φεβ 2014 (πριν από 3 χρόνια και 1 μήνα)

46 εμφανίσεις

Limnol. Oceanogr., 40(3), 1995, 582-588
0 1995, by the American Society of Limnology and Oceancgraphy, Inc.
The influence of lake morphometry on sediment focusing
Jules M. Blais and Jacob Ka&T
Department of Biology, McGill University, 1205 Dr. Penfield, Montreal, Quebec H3A 1Bl
Abstract
Sediment focusing is a process whereby water turbulence moves sedimented material from shallower to
deeper zones of a lake. Sediment focusing occurs in lakes in both the erosional (coarse-grained sediments)
and the transportational zone with the latter characterized by discontinuous sedimentation and resuspension.
The zone of accumulation is diefined as the zone where sedimentation is final (i.e. no resuspension) and where
there is no further sediment focusing. A geochemical tracer (Pb) was used to trace sediment focusing patterns
in 12 lakes of different morphomctries. The area occupied by the zone of accumulation is predicted from
the mean basin slope (a,) with 86% of the variance explained. Only in large, exposed lakes in which turbulence
from waves is more severe are the resulting sediment distributions more erratic. This study is the first to
provide a general model of sc:diment focusing patterns among lakes.
The study of lake-bottom dynamics has received at-
tention in the contexts of sediment, contaminant, and
eutrophication studies. The fact that sediments are major
depositories for contaminants and nutrients has made
them a focus for research in limnology. Also, because
sediments provide a record of past events in lakes, an
understanding of the manner in whllch sediments accu-
mulate is essential for hindcasting. Interpretations of the
historical record in sediments will be confounded if the
record has been obscured by sediment resuspension.
The term sediment focusing was coined by Likens
and Davis ( 1975) to describe the resuspension of sedi-
ments in shallower zones by waves and water currents
with subsequent transport to and settling in the deeper
zones of lakes. Much of the work
on
sediment focusing
has examined sedimentation patterns over glacial time
scales (Likens and Davis 1975; Lehman 1975; Davis and
Ford 1982). This emphasis leaves unstated the relation
between sediment distribution and mere dynamic lake
properties such as mixing depth in the water column,
wave and current shear stress, and sediment cohesiveness.
The purpose of our study is to examine patterns of recent
sedimentation (< 120 yr) in relation to physical variables
such as exposure (circular integral of fetch), slope, and
sediment texture (water content, organic content).
Models relating sediment resuspension to surface waves
and mean water velocity have been generated for selected
lakes (Luettich et al. 1990; Hawley ard Lesht 1992; Aald-
erink et al. 1984). In general, they predict resuspension
where wave shear stress exceeds current shear stress and
implicate waves as the main determinant of sediment
resuspension. In all cases, the study lakes were large and
shallow, thereby emphasizing the importance of surface
Acknowledgments
We thank Joseph Rasmussen for helpful suggestions. Claude
Jean and Douglas Craig helped with held and lab work. The
manuscript was improved by suggestion:; from an anonymous
reviewer.
This project was funded by an NSERC PGS scholarship to
J.M.B. and an FCAR Grant to Yves Prairie and J.K.
Contribution 345 of the McGill University Limnology Re-
search Center.
waves in the entrainment of sediment. Laboratory studies
have shown that sediment resuspension is not only a
function of waves and currents, but also depends on par-
ticle size, water content, and biological mixing (Fukuda
and Lick 1980). The obvious complexity has encouraged
case-by-case studies of sediment resuspension; as a result,
no quantitative generality has emerged. For example, Hil-
ton et al. (1986) listed the many processes that have been
attributed to sediment focusing in the past and noted that
none was sufficient to explain much of the variance in
their study lake.
Hakanson (1977) divided lake bottoms into three zones,
based on differences in their potential for resuspension.
These are the erosional zone, transportation zone, and
accumulation zone. The zone of erosion is marked by
coarse-grained, noncohesive sediments and is found in
areas of high water turbulence. Rowan et al. (1992) dc-
veloped a model that predicted the extent of the zone of
erosion using exposure as a surrogate for wave energy and
underwater slope. Beyond this zone lies the zone of trans-
portation which is a zone of discontinuous sedimentation,
where sediment accumulation is interrupted by infre-
quent periods of resuspension and associated transport
during overturn or storm events. Below the zone of trans-
portation lies the zone of accumulation which is marked
as the area where no further focusing takes place.
Ac-
cording to Hakansons scheme, the 50% water content of
surficial sediments marks the transition between the zones
of erosion and transportation, whereas the 75% water
content of surficial sediments marks the transition be-
tween the zones of transportation and accumulation. Here
we use Pb as a geochemical measure of sediment depo-
sition rather than sediment texture to define the zones of
sediment distribution. In addition, we model sediment
redistribution in lakes by relating the area occupied by
the accumulation zone with lake morphometric variables.
Methods
We used Pb as a geochemical tracer of sediment de-
position because it is well suited in that almost all of the
Pb in sediments is derived from anthropogenic inputs by
direct deposition on lake surfaces (Dillon and Evans 1982;
582
Sediment focusing in lakes
583
Blais and Kalff 1993). Inputs of Pb from catchments are
virtually negligible (Schut et al. 1986; Blais and Kalff
1993) as is the loss of Pb through the outflow of lakes
with water residence times > 1 yr (Schut et al. 1986).
Therefore, distributions of Pb should reflect distributions
of sediment deposition without being influenced by fac-
tors such as terrestrial inputs or loss through the outflow.
The sediment profiles of Pb are characterized by a sharp
surface peak that is attributed to increases in coal com-
bustion as well as mining and smelting activities in the
late 19th century, further enhanced by the combustion of
leaded gasoline which began in the 1920s (Nriagu 1990).
The onset of this Pb peak in sediments, when compared
with 210Pb-dated cores for lakes in southern Quebec, cor-
responds to 1886+ 15 yr (95% C.L., Blais et al. 1995). Pb
deposition at a given site (in rug Pb cm-2) is correlated
with the sedimentation rate in lakes (Evans 1980). How-
ever, this may not be universally valid. Differential par-
titioning of Pb onto different sediment fractions may re-
sult in discrepancies between bulk sedimentation rates
and Pb deposition. Here we use Pb deposition values
derived from 120 cores to quantify the distribution pat-
terns of Pb across a variety of lakes.
Sample collection-
Sediment cores were extracted along
transects from eight lakes in the Eastern Townships and
two lakes in the Laurentians of Quebec (Table 1) with a
modified K-B gravity corer. The locations and specifics
of these lakes are described elsewhere (Blais and Kalff
1993). At each site, depth was recorded, and slope was
derived by echo sounding perpendicular to basin con-
tours. Sediment was extruded on shore with a vertical
extrusion device. Cores were sliced into 1 -cm horizontal
slices for the top 10 cm and into 2-cm slices thereafter.
Sediment samples were stored in clean, preweighed poly-
ethylene vials. In the lab, samples were weighed and oven-
dried at 60-80°C. Subsamples were ashed at 550°C to
derive the organic content of the sediments (Dean 1974).
Analytical procedures - The concentration of Pb in sed-
iment samples was determined by digesting 0.5 g of dry,
crushed sediment in dilute aqua regia (3 HCl : 3 Hz0 :
HN03) and measuring Pb levels with a flame atomic
absorption spectrometer (Perkin Elmer 3 100). Standard
reference material (NBS No. 1572) was digested and an-
alyzcd to obtain the extraction efficiency and to detect
temporal variations in the analytical effectiveness. The
extraction efficiency of Pb was 100%; extraction repro-
ducibility was within lo%, well within the limits set by
the National Bureau of Standards.
Statistical analysis
-Typical sediment Pb profiles from
two of the study lakes are shown in Fig. 1. Concentration
declines with depth until a stable background Pb con-
centration is reached. The anthropogenically derived Pb
can therefore be quantified in sediment cores according
to the equation (Evans 1980):
pbA = Wb, -APbl DW.
c
(1)
Table 1.
Basin morphometric variables used in this analysis.
afP
is
mean basin slope (o/o).
A,
is lake area (km2), Z,,,, is mean
depth, Z,,,
is maximum depth (m), and the dynamic ratio is
defined as
A,j2 : Z,,,,.
DY-
z *
mean -
namic
Lake
4J
4 Zman Znax Znax
ratio
1
Type 1
Aylmer
2.8 8.05 8.5 36.2 0.23
0.33
Brompton 4.9
1.95
12.5 42.3 0.29
0.11
Bowker
4.9 2.28 25.9 59.0 0.44
0.06
Massawippi 7.2
17.9 41.6 85.7 0.48
0.10
Brome 1.1 14.5 5.8
12.8 0.45
0.65
Type 2*
Cromwell
11.1
0.04 3.0 9.0 0.33
0.07
Crochc 17
0.21
4.4 10.5 0.42
0.10
Lovering
5.0
4.62
10.3 24.9 0.41
0.21
Nicolet
14.8 3.97 17.3 41.5 0.42
0.11
Bob-/
18.0 2.28
17.8 65
0.27 0.09
Red Chalkt 15.8
0.44 16.7 38
0.44
0.04
Costellot
13.0 0.34
8.4 17.5 0.48
0.07
* Lake Or-ford is omitted.
j- From Evans 1980.
PbA is the anthropogenic Pb burden (pg cm-2), Pbj the
Pb concentration at the ith horizontal section of a core
6-e g-9, Pb, th
e ac ground Pb concentration derived
b k
at depth in the core below the anthropogenically influ-
enced layer (pg g-l); DW the dry weight of ith section (g)
and A, the area of, core (cm2).
The distributions of anthropogenic lead (Pb,) in sed-
iments were used to characterize the sedimentation zones
of transportation (ZT) and accumulation (ZA). This char-
acterization was accomplished by devising a method to
determine the depth at which PbA reaches its maximum
(i.e. where no further focusing takes place) and defining
this depth as the transition from ZT to ZA.
The depth of transition from ZT to ZA was determined
statistically by iterative regression. The purpose of this
procedure is to determine the depth in the lake at which
PbA increases no further. The data points are added to
the Pb-depth plot for each individual lake in ranking
order of descending depth until plj of the model,
PbA = @liZ + Ci + &i,
(2)
is significant at the level a! = 0.05, where Z is local depth
pli is the slope term, Ci the y-intercept, and Ei the error
term for lake i. This depth represents the zone at which
PbA begins to decline when moving from right to left along
the Pb,-depth plot, marked with a dashed line and de-
noted by a  1 in the figures. Next, data points are added
sequentially in ranking order of ascending depth until @2i
of the model,
PbA = /3liZ + /32iz2 + Ci + &I,
(3)
is significant at the level cy = 0.2. This depth represents
the point of curvature when moving from left to right
584
Blais and Ku&?
PI3 concentration (ppm)
20
100
200 300
Fig. 1. Typical Pb concentration profiles in sediment cores.
Note the surface concentration peak corresponding to industrial
activity and the decline with depth until a stable background is
reached.
along the Pb,-depth distribution and is marked by a line
dencted with a 2 in the figures. An cy level of 0.2 was
chosen to sensitize the model to detect curvature, thereby
minimizing the overlap between lines 1 and 2. If & never
reached a significance level of 0.2, then line 2 was drawn
at z,,,.
We define
tlhe
transition from ZT to ZA as the
midpoint between lines 1 and 2 (solid line).
Blzsin morphometry- Basin morphometric variables
were derived from biathymetric maps. Lake area was de-
rived by planimetry. Lake and contour perimeters were
obtained with a rotometer. Three different measures of
slope are used in this, analysis. Mean basin slope between
contours (CXJ was used to quantify sloping patterns across
each basin and was obtained with the equation (Hakanson
1981)
GYP q
= (II + 12) LJ20A.
(4)
I, and 1, are the lengths (km) of the two contour lines
being considered, L, is the change in depth between con-
tours (m), and A is the area between the two contour
lines (km2). Mean slope for the entire lake area (a,) was
used as the principal predictor of whole-lake sediment
focusing patterns, olbtained as follows (Hakanson 198 1):
I
a p = (Z,,/2 + I, + I, + . . . + !,_I
+ 1,/2)2,,,/ 1 OnA,.
(5)
lj are the lengths of the contour lines (km), lo is the shore-
line: length (km), Z,,,
the maximum depth of the lake
(m), n the number ofcontour lines, and AL the lake surface
area (km2). Lake variables are shown in Table 1 (except
Orford). Finally, site-specific slope at each coring site was
obtained by depth sounding perpendicular to basin con-
tours and calculating slope as rise over the run.
Sediment characterization -The sediments were clas-
sified into zones of erosion, transportation, and accu-
mulation. The zone of erosion (ZE) is defined here in the
same way as by Hakanson ( 1977) as sediments with water
content ~50%. ZT extends from this point to the depth
where Pb increases no further. ZA is defined simply as
the area below ZT.
Results
All study lakes, except Orford, exhibited a focusing of
sediment material, as demonstrated by the increase in
anthropogenic Pb (Pb,) with depth (Figs. 2-3). One fea-
ture that was evident in some of the lakes was that the
Pb-depth curves formed an asymptote with the plateau
generally occurring at a depth of -20 m. The statistical
significance of this nonlinearity was tested by fitting both
a linear model to the distribution of the type
PbA = pi Z + Ci + &i
(6)
as well as a nonlinear model of the type
PbA = Ki[ 1 - exp( -Z)] + Ci + &i
(7)
where /3i, Kj, and Ci are constants. Next, an F-test was
used to determine whether the nonlinear model had a
significantly better fit than the linear model. Significant
nonlinearity (as determined by F-test) was observed for
lakes Massawippi, Brompton, Aylmer, Brome, and Bow-
ker. The plateau distribution shows that there is a thresh-
old depth beyond which sedimented material is no longer
resuspended and redeposited. Because Lake Orford ex-
hibited no significant correlation between PbA and depth,
it could not be analyzed by iterative regression in the
subsequent analyses. Thus it was removed.
Type I focusing-The focusing pattern in the five lakes
with significant nonlinearity, as determined by F-test in
the Pb-depth distributions listed above (hereafter referred
to as type 1 focusing) characterizes lakes with shallow
basin slopes (mean basin slope, 4.2%). The pattern ob-
served is compared to a sediment parameter (water con-
tent) in Fig. 2. Lakes Brompton and Massawippi exhibit
a transition from coarse-grained, noncohesive sediments
(erosional environments where water content is ~50%)
to fine-grained, cohesive sediments (depositional envi-
ronments where water content is > 75%, Hakanson 1977).
The transition is less apparent in Lakes Aylmer and Bow-
ker.
The distribution of PbA in Lake Brompton is mirrored
precisely in the distribution of sediment water content
(Fig. 2), with the plateau of PbA deposition occurring at
the transition to a sediment depositional environment.
In Lake Aylmer, the increase in PbA accumulation ends
at a depth of 18 m (Fig. 2) but the sediment water content
shows that the transition to a depositional environment
Sediment focusing in lakes
585
Brome
Brompton
Aylmer
Bowker
I
TI .;* I.
lIIl
E. i
I
1
2
Massawippi
5
10 15
10 20 30 40 so
10 20 30 40
20 40 60 80
20 40 60 80
depth (m)
Fig. 2. Focusing patterns plotted against depth for lakes showing type 1 focusing. Water content is plotted against depth for
comparison. The zones defined as erosion (E), transportation (T), and accumulation (A) are shown with bars. Dashed lines are
from iterative regressions from Eq. 2 and 3, and the transition bctwecn T and A is shown with solid lines.
occurs at a depth of
-7 m. The observation that sedi-
mcnts
in Lake
Aylmer focus beyond the sediment tran-
sition zone is
probably attributable to greater slopes (up
to 13% at a depth of 17 m which could coincide with
sediment slumping (Hakanson 1977) and thereby in-
crease sediment focusing. The PbA plateau in Lake Bow-
ker also occurred deeper than the transition to cohesive
sediments. This is likely the result of high slopes in the
first 20 m. Lake Brome is a large (14.5 km2), shallow
Glm =
5.8 m) lake where the sediments are likely af-
fected by turbulence from waves. Therefore the PbA dis-
tributions in Brome are more variable and the scatter of
PbA in Fig. 2 is considerable. Lastly, in Lake Massawippi,
PbA shows a similar distribution as the sediment water
content, demonstrating that there is no resuspension of
material below the textural boundary.
Type 2 focusing-The second pattern observed is one
in which focusing is a linear function of depth. The pattern
characterizes lakes with steep slopes like in Cromwell,
Creche, Nicolet, Lovering, and Orford. The sample was
augmented by including published PbA and water content
data for Lakes Costello, Red Chalk, and Bob from Evans
(1980). In contrast to type 1 sediment focusing patterns,
type 2 focusing does not demonstrate an asymptote in
the Pb-depth plot, although there is a clear transition in
sediments toward a depositional environment as shown
by a plateau in sediment water content (Fig. 3). Type 2
lakes have characteristically high slopes (mean CY~ is 13.5%,
compared to 4.2% for type 1 lakes), thus suggesting that
type 2 focusing is driven by transport of material along
sloping gradients. Because the proportion of the lake bot-
tom occupied by the zone of accumulation is not repre-
sented by the Pb-depth curves, we used hypsographic data
to asses the proportional areas of the sedimentation zones
in the following analysis of the size of each of the three
zones in lakes.
Sedimentation zones-The present study demonstrates
the need to re-evaluate the sediment classification by
Hakanson (1977, 1982). Hakanson defined the zone of
accumulation by the transition to soft sediment (water
content >75%). Yet, in lakes showing a type 2 focusing
pattern, it is clear that Pb deposition continues to increase
at depths below this transition to soft sediment. Sediment
water content, therefore, does not denote the boundary
of the transportation zone. These observations indicate
that Hakanson probably overestimated the boundaries of
the accumulation zone in his earlier studies.
The percent area occupied by the accumulation zone
(hereafter %ZA) was determined for each lake by calcu-
lating the area occupied by the region below the depth of
the transition from the transportation to accumulation
zones as a percentage of the total lake area. We compared
the %ZA zone with mean basin slope (CU,) in Fig. 4. The
model is as follows:
%ZA = 49.92(+3.73) - 2.5O(LO.31) c&.
(8)
alp is the mean basin slope as calculated from Eq. 5. The
model yielded an r2 =
0.86,
SE,,, = 6.81, P <O.OOOl, n
= 12. Standard errors of parameter estimates are in pa-
rentheses. There is a negative correlation between %ZA
and cy, indicating that the area occupied by ZA is largely
a function of the mean slope of the entire basin. On an
individual lake basis, neither site slope nor mean basin
slope between contours were significant predictors of PbA.
Discussion
The zones of accumulation, erosion, and transportation
have traditionally been identified by using a measure of
sediment texture such as water content. Thus it would
appear that the accumulation zone as defined on the basis
of lead is more restricted (and the transportation zone
more extensive) than those determined from previous
586
Blais and Kalf
Cromwell Creche
Nicolet
Lovering
Orford
8 70
3
50
.
.
. . . . l . .
Lr-7
2 4 6 8 10
6 10 14
10 20 30 40
10 20 30
10 20 30 40 50
depth (m)
Fig. 3. As Fig. 2, but for lakes showing a type 2 focusing pattern.
studies based on sedirnent texture. None of the lakes in
our study should have accumulation areas (O/oZA) < 60%
according to the model of Hakanson (1982). Yet, some
of the lakes have a %ZA < 5% as it is defined in this study.
The discrepancy is the result of an overestimation of the
accumulation zone and an associated underestimation of
the transportation zone due to the sediment classification
scheme of the earlier work which used water content to
assess sediment accumulation.
The single best determinant of the transition zone be-
tween transportation and accumulation is mean basin
slope. Correlates of water turbulence such as exposure
(circular integral of fetch, Rasmussen 1988) or the dy-
namic ratio (ALh/Zmeiln)
were not significant. Hakanson
(1982) documented a strong influence of the dynamic
6o1
I
I I
t
l
Brome
50
40
20
20
10 15
mean slope (%)
Fig, 4. Pb accumulation area plotted against mean basin
slope for the 12 study lakes. Asterisks denote lakes from Evans
1980.
ratio on sediment distribution patterns in a data set con-
taining large, but relatively shallow lakes characterized
by a much larger dynamic ratio (0.06-5.25, mean = 1.9 1)
compared to the lakes in our study (0.04-0.65, mean =
0.16) which are more typical of lakes in this region.
Slope impacts sediment distributions in several ways.
First of all, slope has been implicated as an important
determinant of turbidity currents that effectively trans-
port sediment material to deeper zones in lake basins.
Furthermore, sediment material may slide down steep
gradients in response to turbulence from earthquakes,
overturns, or storm events.
The extent of focusing is of great importance in studies
concerned with sediment contaminant distribution, as
well as for paleolimnological reconstructions. Sediment
resuspension will have a profound effect on the cycling
and distribution of both nutrients and pollutants. Focus-
ing results in an enhanced accumulation fine-grained par-
ticles in the ZA. These particles provide a large surface
area for the sorption of heavy metals, organic pollutants,
and nutrients such as N and P. The process of sediment
focusing may enhance uptake of contaminants by the
benthic biota and internal loading of P and heavy metals
from the sediments during periods of bottom-water an-
oxia. Evans (1980) found surface sediment concentrations
of Pb to range between 9 and 290 ppm within a single
lake, with the high values the result of sediment focusing
of fine materials toward the center. Differences in sub-
strate metal concentration of this magnitude have been
shown to impact body burdens of benthic invertebrates
and fish (Anderson et al. 1978). In addition, benthic feed-
ers such as Mysis, Pontoporca, and Chaoborus, which
ingest sediments during the day at the deepest part of the
lake, transport sedimented material to the water column
at night through feces or molted exoskeletons as part of
their diel vertical migration (Van Duyn-Henderson and
Lasenby 1986), thus acting as a possible vector for con-
taminated deep sediments to the water column. Internal
loading processes involving redox-sensitive elements (such
as Fe, Mn, and P) involve a flux of these constituents to
Sediment focusing in lakes
587
the water column as the 0.2 isovolt moves above the
sediment-water interface (Mortimer 194 1). Consequent-
ly, internal loading of these elements from sediments may
be enhanced if surficial concentrations in sediments are
elevated through focusing of fine materials.
It is widely accepted that deep lakes exhibit the highest
degree of sediment focusing (i.e. Evans and Rigler 1985;
Evans 199 1). Our results indicate that this is not true. In
fact, many of the deepest lakes demonstrated the least
. amount of sediment focusing (e.g. Lakes Massawippi,.
Brompton, and Bowker). The notion stems from the fact
that earlier studies of sediment focusing used either small,
deep lakes that were characterized by steeply sloping ba-
sins and consequently had a relatively large transporta-
tion zone or examined large, shallow lakes that were af-
fected greatly by turbulence from wind. The observation
that focusing is a function of lake form rather than depth
corroborates the findings of Likens and Davis (1975), who
reported that there was an apparent decrease in sedi-
mcntation rate as the sediments of Mirror Lake became
more recent. Conceptually, one can imagine that a lake
will exhibit a high degree of sediment focusing early in
its ontogeny when it has a steeper basin, and this focusing
will decrease as the basin fills, thereby widening the ac-
cumulation zone and causing an apparent decrease in the
sedimentation rate of recent sediments. The present anal-
ysis, which demonstrated a tight link between the accu-
mulation zone and basin slope, makes this plausible. It
is also interesting to note that many of the lakes with
relatively large accumulation zones (Massawippi, Bromp-
ton, Bowker, and Aylmer) were under the Champlain Sea
during the last glaciation and therefore have received more
sediment than lakes on the Precambrian Shield which
were not part of the Champlain Sea (i.e. Lakes Cromwell,
Creche, Red Chalk, Costello, and Bob).
It is important for a variety of studies to assess the area
of lakes occupied by the zones of erosion, transportation,
and accumulation. This classification is especially im-
portant for studies requiring multiple cores to assess whole-
lake sediment flux. Efforts are presently being made to
maximize core sampling efficiency, using the sediment
classification of Hakanson (1977) and Rowan et al. (1992)
as a guide (Rowan et al. in prep.). Rowan et al. (in prep.)
estimate that only 5-10 cores are required to obtain a 5%
level of precision in the rate of mass sediment accumu-
lation in the zone of accumulation. The efficiency of sam-
pling regimes would be enhanced by assessing the extent
of the zone of accumulation as defined here. By calcu-
lating %ZA using Eq. 8, one can arrange a sampling pro-
gram where no more than 5-l 0 cores are required from
this region in order to obtain rates of mass accumulation
that are within 5% of the true mean.
References
AaDERINK, R. H., L. LIJKLEMA, J. BREUKELMAN, W.
VAN
RAAPHORST, AND
A. G.
BRINKMAN. 1984. Quantification
of wind induced resuspension
in a shallow lake. Water Sci.
Technol. 17: 903-9 14.
ANDERSON, R. V., W. S. VINIKOUR, AND J. E. BROWER. 1978.
The distribution of Cd, Cu, Pb, and Zn in the biota of two
freshwater sites with different trace metal inputs. Holarct.
Ecol. 1: 377-384.
BLAIS, J. M., AND J. KALFF. 1993. Atmospheric loading of Zn,
Cu, Ni, Cr, and Pb to lake sediments: The role of catchment,
lake morphometry and physico-chemical properties of the
elements. Biogeochemistry 23: l-22.
- - R.J. CORNETT,AND R.D. EVANS. 1995. Eval-
uaiion of ;lPb dating in lake sediments using stable Pb,
Ambrosia
pollen
13Cs. J. Paleolimnol. In press.
DAVIS, M. B., AND M. S. FORD. 1982. Sediment focusing in
Mirror Lake, New Hampshire. Limnol. Oceanogr. 27: 137-
150.
DEAN, W. E., JR. 1974. Determination of carbonate and or-
ganic matter in calcareous sediments and sedimentary rocks
by loss on ignition: Comparison with other methods. J.
Sediment. Petrol. 44: 242-248.
DILLON, P. J., AND R. D. EVANS.
1982. Whole lake lead bur-
dens in sediments of lakes in southern Ontario, Canada.
Hydrobiologia 91: 12 l-l 30.
EVANS, R. D. 1980. Measurement of sediment accumulation
and phosphorus retention using Pb-2 10 dating. Ph.D. the-
sis, McGill Univ. 169 p.
- 199 1. The impact of sediment focussing on total re-
sidual 210Pb: Implications for choice of a dating model. Int.
Ver. Theor. Angew. Limnol. Verh. 24: 2335-2339.
- AND F. H. RIGLER. 1985. Long distance transport of
an;hropogenic lead as measured by lake sediments. Water
Air Soil Pollut. 24: 14 l-l 5 1.
FUKUDA, M. K., AND W. J. LICK. 1980. The entrainment of
cohesive sediments in freshwater. J. Geophys. Res. 85: 28 13-
2824.
HAKANSON, L. 1977. The influence of wind, fetch and water
depth on the distribution of sediments in Lake Vanern,
Sweden. Can. J. Earth Sci. 14: 397-412.
~ . 198 1. A manual of lake morphometry. Springer.
- 1982. Lake bottom dynamics and morphometry: The
dynamic ratio. Water Resour. Res. 18: 1444-1450.
HAWLEY, N., AND B. M. LESHT. 1992. Sediment resuspension
in Lake St. Clair. Limnol. Oceanogr. 37: 1720-1737.
HILTON, J.,J.P. LISHMAN, AND P.V. ALLEN. 1986. Thedom-
inant processes of sediment distribution and focusing in a
small, eutrophic, monomictic lake. Limnol. Oceanogr. 31:
125-133.
LEHMAN, J. T. 1975. Reconstructing the rate of accumulation
of lake sediment: The effect of sediment focussing. Quat.
Res. 5: 541-550.
LIKENS, G. E., AND M. B. DAVIS.
1975. Post-glacial history of
Mirror Lake and its watershed in New Hampshire, U.S.A.,
and initial report. Int. Ver. Theor. Angew. Limnol. Verh.
19:
982-993.
LUETTICH, R. A., JR., D. R. F. HARLEMAN, AND L. SOMY~DY.
1990. Dynamic behavior of suspended sediment concen-
trations in a shallow lake perturbed by wind events. Limnol.
Oceanogr. 35: 1050-1067.
MORTIMER, C. H. 194 1. The exchange of dissolved substances
between mud and water in lakes. J. Ecol. 29: 208-329.
NRIAGU, J. 0. 1990. The rise and fall of leaded gasoline. Sci.
Total Environ. 92: 13-28.
RASMUSSEN, J. B. 1988. Littoral zoobenthic biomass in lakes
and its relationship to physical, chemical, and trophic fac-
tors. Can. J. Fish. Aquat. Sci. 44: 990-1001.
ROWAN, D. J., J. KALFF, AND J. B. RASMUSSEN. 1992. Esti-
mating the mud deposition boundary depth in lakes from
wave
theory. Can. J. Fish. Aquat. Sci. 49: 2490-2497.
588
Blah and Ku@
SCHUT,
P. H., R. D. EVANS,
AND
W. A.
SCHEIDER.
1986. Vari-
ation in trace metal exports from small Canadian shield
watersheds. Water Air Soil Pollut. 28: 225-237.
VAN DUYN-HENDERSON,
J. A.,
AND
D. C.
LASENBY.
1986. Zinc
and cadmium transport by the vertically migrating opos-
sum shrimp
Mysis dicta.
Can. J. Fish. Aquat. Sci. 43:
1726-1732.
Submitted: 12 October
1993
Accepted: 5 October 1994
Amended: 25 January 1995