ARTICLE IN PRESS
Continental Shelf Research 23 (2003) 1201–1225
Sediment source strength, transport pathways and
accumulation patterns on the Siberian–Arctic’s Chukchi
and Laptev shelves
C. Viscosi-Shirleya,*, N. Pisiasa, K. Mammoneb
a
College of Oceanic and Atmospheric Sciences, Oregon State University, 104 Ocean Admin. Bldg., Corvallis, OR 97331-5503, USA
b
Hewlett-Packard Company, 1000 NE Circle Blvd. Corvallis, OR-97330, USA
Received 18 July 2002; accepted 4 April 2003
Abstract
In this study, we estimate sediment source strength and determine sediment transport/accumulation patterns on the
Chukchi and Laptev shelves for lithogenic material from several key source regions. In the western Laptev Sea,
sediments from the Siberian flood basalt province account for o20% of surface sediments, while detritus from the
eastern Laptev’s Lena and Yana Rivers constitutes as much as 40% of seafloor sediments. Eastern Laptev Sea
sediments similarly reflect inputs from multiple sources; here, however, local inputs from the Lena and Yana Rivers are
most prevalent. Chukchi Sea sediments are more homogeneous in composition, with sediments originating from the
Okhotsk-Chukotsk volcanic belt and Bering Strait inflow comprising over 60% of surface sediments except near
Wrangel Island and along sections of the Siberian coast. In the Chukchi portion of our study area, the dominant
sediment dispersal pathways are from the Bering Strait region to the north and from Long Strait to the east, into the
central Chukchi Sea. There appears to be comparatively minimal sediment transport parallel to the Siberian coast at
our sample sites (>40 m water depth). In contrast, at our Laptev Sea sites (typically >20 m water depth) sediments
supplied by the Lena and Yana Rivers move primarily eastward/northeastward parallel to the coast, with relatively
minimal transport to the west/northwest or north to the central shelf. Good correspondence between sediment
accumulation patterns and currents on the Chukchi shelf indicates water circulation is an important sediment transport
mechanism in this marginal sea. In the Laptev Sea, a combination of river outflow, cyclonic water circulation and the
Siberian Coastal Current controls sediment distribution. We speculate that sediment ice rafting may also affect shelf
sedimentation patterns through a combination of factors. Our findings have implications for the fate of particle reactive
contaminants released to the Siberian continental margin.
r 2003 Elsevier Ltd. All rights reserved.
Keywords: Accumulation (of sediment); Arctic; Chukchi and Laptev Seas; River outflow; Sediment chemistry; Sediment sources;
Sediment transport; Siberia and Alaska; Water currents
1. Introduction
*Corresponding author.
E-mail address: [email protected] (C. Viscosi-Shirley).
Making up over one-third the total area of the
Arctic continental shelf (Silverberg, 1972; Holmes,
1975; Macdonald et al., 1998), the Siberian shelf
0278-4343/03/$ - see front matter r 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0278-4343(03)00090-6
ARTICLE IN PRESS
1202
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
receives large amounts of sediment from river
discharge (Gordeev et al., 1996; Lisitzin, 1996;
Are, 1999) and coastal erosion (Timokhov, 1994;
Are, 1999; Rachold et al., 2000). Recent studies
indicate significant potential exists for polluting
the shelf with various particle reactive substances
including heavy metals, organochlorines and, to
a lesser extent, radionuclides (Macdonald and
Bewers, 1996; Pfirman et al., 1997; AMAP,
2002). Determining the dominant sediment transport directions and accumulation patterns of
Siberian shelf sediments thus has important
implications for identifying regions most threatened by particle reactive pollutants. On a more
global scale, characterizing modern sedimentation
patterns and the factors controlling them also
provides a necessary baseline for interpreting the
record of paleoclimatic change preserved in Arctic
sediments (Kassens and Thiede, 1994; Stein and
Korolev, 1994). Such studies are critical in light of
growing evidence that the Arctic is a key part of
the global climate system, which will both respond
sensitively to and amplify the effects of global
climate change (Walsh, 1991).
Sediment dispersal and accumulation patterns
reflect the net effect of spatial and temporal
changes in sediment fluxes to the shelf and
sediment transport processes. Since sediment
transport agents such as currents and ice rafting
are highly spatially and temporally variable on the
Siberian shelf, we cannot predict sedimentation
patterns in this region a priori on the basis of
physical oceanographic data. A number of studies
have characterized the clay and heavy mineral
composition of various Siberian shelf sediment
sources and, to a limited extent, their sediment
chemistry (Naidu et al., 1982; Nurnberg
.
et al.,
1994; Behrends et al., 1999; Eisenhauer et al., 1999;
Rachold, 1999; Rossak et al., 1999; Wahsner et al.,
1999; Schoster et al., 2000). These data provide
direct evidence of sediment movement along
various trajectories throughout the shelf and from
there into the central Arctic basin. However,
because the compositional data in these studies
are expressed as abundances (e.g. ppm, %, etc.)
rather than fluxes (e.g. mg/cm2/yr), the distribution of material from a particular source may be
affected by dilution from other sources. This
prevents comparing the magnitude of sediment
transport along these trajectories and thus the
accurate identification of dominant sediment dispersal pathways.
In this study, we determine fluxes to the Siberian
shelf of sediment from key source regions based on
of measured surface sediment geochemistry and
published sedimentation rates. Previously (ViscosiShirley et al., 2003), we described the major and
minor element geochemistry of 81 surface sediment samples taken from Chukchi, East Siberian
and Laptev sediment cores collected by the USCG
in the 1960s (Figs. 1a and b). On the basis of these
data we demonstrated that various geologic
terrains supplying sediment to the Siberian shelf
have distinct geochemical signatures. Here we
estimate the percent abundance of these geochemical endmembers in the 81 surface sediment
samples. We use these endmember percentages to
assess the relative strength of individual sediment
sources on various parts of the shelf. For the
Laptev and Chukchi Seas, we compile published
mass accumulation rate data and linear sedimentation rate data, which we convert to mass
accumulation rates. Fluxes of the sedimentary
geochemical endmembers are estimated based on
the calculated endmember abundances and published sedimentation rate data. Dominant sediment transport pathways for sediments
accumulating within our study area are inferred
from spatial variability in endmember fluxes.
Finally, to the extent possible given available
physical oceanographic data, we evaluate controls
on observed sediment accumulation patterns.
2. Background
2.1. Sedimentary geochemical endmembers
Sediment geochemistry can differentiate detritus
from various sediment sources on the Siberian
shelf. In Viscosi-Shirley et al. (2003), we characterized the multi-element chemistry (Si, Al, K,
Mg, Sr, La, Ce, Nd) of 81 core top samples taken
from sediment cores collected in the Laptev, East
Siberian and Chukchi Seas by the USCG between
1962 and 1964 (Fig. 1b). In order to discern
ARTICLE IN PRESS
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
1203
(a)
SIBERIAN
PLATFORM
CT
CHUKOTKA TERRAIN
(CT)
VERKHOYANSK
MOUNTAINS
VB
KOLYMA-OMOLON SUPERTERRAIN
OKHOTSK-CHUKOTSK
VOLCANIC BELT
(VB)
Fig. 1. (a) Arctic Ocean and marginal seas. (b) Siberian-shelf surface-sediment sample location () and geology map.
compositional trends in the lithogenic fraction of
shelf sediments, we expressed the bulk sedimentary
chemical data on an organic (biogenic opal,
carbonate and organic matter) free basis. A
combination of Q-mode factor analysis and x–y
scatter plots of these data indicated there are four
distinct, endmember geochemical compositions
evident in Siberian shelf sediments (Table 1). Each
of these geochemical endmembers reflects sediment input from a different geologic terrain
(Fig. 1b). (1) The shale endmember (Al, K and
rare-earth element rich sediment) is eroded from
fine-grained marine sedimentary rocks of the
Verkhoyansk Mountains and Kolyma-Omolon
superterrain, and discharged to the shelf by the
Lena, Yana, Indigirka and Kolyma Rivers. (2)
The basalt endmember (Mg rich) is supplied by
NE Siberia’s Okhotsk-Chukotsk volcanic belt and
Bering Strait inflow, and is prevalent in Chukchi
Sea sediments. (3) The mature sandstone
ARTICLE IN PRESS
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
1204
Table 1
Sedimentary endmember characteristics, and least-squares-regression partitioning parameters
Endmember
Shale
Basalt
Mature
sandstone
Immature
sandstone
Sample IDa
Latitude
Longitude
NW63-119
72.80
133.00
NW63-29
69.89
174.44
NW63-34
69.32
177.58
NW63-153
74.53
125.93
Si (wt%)
Al
K
Mg
Sr (ppm)
La
Ce
Nd
27
9.5
2.9
1.7
148
42
92
38
30
7.5
2.1
1.7
194
26
50
23
38
4.6
1.8
0.6
145
21
43
18
34
6.5
2.5
0.7
281
30
63
25
Sand (wt%)
Silt
Clay
o1
51
49
3
51
45
61
19
20
58
18
24
a
Weights
Coefficients of
v=variable value determination
in each spl.
% explained
1/sqrt(v)
5/sqrt(v)
10/sqrt(v)
10/sqrt(v)
1/sqrt(v)
1/sqrt(v)
1/sqrt(v)
1/sqrt(v)
81
81
66
87
100
91
95
91
Compositionally extreme samples from within the geochemical data set were selected to represent each endmember.
endmember (Si rich) is present proximal to
Wrangel Island and sections of the Chukchi Sea’s
Siberian coast and is derived from the sedimentary
Chukotka terrain found in these areas. (4) The
immature sandstone endmember (Sr rich) is
abundant in the New Siberian Island region and
reflects inputs from sedimentary rocks that comprise the islands. The immature sandstone endmember is also prevalent in the western Laptev
Sea, where it is eroded from sedimentary deposits
blanketing the Siberian platform that are compositionally similar to those on the New Siberian
Islands. Western Laptev can be distinguished from
New Siberian Island region sediments by the
presence of the basalt endmember, which indicates
Siberian platform flood basalts are another source
of western Laptev sediments. We also found
evidence that the mature and immature sandstone
endmembers, which consist of much coarser
sediment than the other endmembers (Table 1),
reflect erosional processes as well as provenance.
In this study, we consequently focus on the
distributions of shale and basalt endmember fluxes
to evaluate sediment dispersal patterns.
2.2. Physical oceanography
On the Siberian shelf, sediment transport agents
such as currents and sea ice exhibit high spatial
and temporal variability. In the region of the
Chukchi Sea spanned by our samples, currents at
all water depths typically flow northward following bottom contours, although they can exhibit
substantial short-term wind-driven fluctuations
(Fig. 2) (Coachman et al., 1975; Coachman and
Shigaev, 1992; Roach et al., 1995; Weingartner
et al., 1998a). This cross-shelf flow passes around
Herald Shoal to exit the region via Hope Sea
Valley and Barrow Canyon. Laptev and East
Siberian currents are believed to flow in cyclonic
gyres both at the surface and at depth (Dmitrenko
et al., 1995; Hass et al., 1995; Timokhov, 1994;
Pavlov et al., 1996) and thus also have northerly,
offshore components, strongest in the eastern part
of each sea. The Siberian Coastal Current flows
alongshore toward the east from the Laptev to the
Chukchi Sea, where it turns north–northeast to
join the cross-shelf flow (Coachman and Shigaev,
1992; Weingartner et al., 1998b). Studying the
ARTICLE IN PRESS
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
1205
75
˚
Laptev
Sea
New
Siberian Ils.
˚
0m
20
East Siberian Sea
m
100
50m
Chukchi Sea
Herald Sea
Valley
Herald
Wrangel I.
65
Shl. 40m
Alaska
SCC
˚
m
0˚
Be
11
40
rin
20
m
gS
tr.
Yana R.
Indigirka R.
0˚
60
Lena R.
-1
˚
12
Barrow Canyo
n
70
m
00
10
0˚
-17
Kolyma R.
130
˚
˚
180
Siberia
140˚
170˚
150˚
160˚
Fig. 2. Schematic of Siberian shelf circulation. See text for discussion of spatial and temporal variability in shelf currents.
SCC=Siberian Coastal Current.
Siberian Coastal Current on the Chukchi shelf
using fall measurements made between 1992 and
1995, Weingartner et al. (1998b) found that in fall
of 1995 the easterly flowing Siberian Coastal
Current failed to develop and coastal flow was to
the west–northwest. According to Pavlov (1995)
and Munchow et al. (1998), similar flow reversals
have been observed in the Laptev and East
Siberian Seas. Although it appears alongshore
flow is generally to the east on the basis of
Weingartner et al.’s work, we have yet to clearly
establish the frequency and duration of these flow
reversals, and the mechanisms that drive them.
While current velocity data are sparse, available
information suggests the most vigorous current in
our study area is the Chukchi’s mean offshore
northerly flow, with velocities on the order of
25 cm/s in Bering Strait decreasing to B10 cm/s in
the central Chukchi (Weingartner et al., 1998a).
Maximum flow velocities along this offshore
gradient range from 100 cm/s in the strait to
40 cm/s in the central Chukchi. Current velocities
in the Laptev and East Siberian gyres appear to be
much slower, typically o10 cm/s, although speeds
up to 25–50 cm/s have been observed (Holmes and
Creager, 1974; Munchow et al., 1998). Alongshore, data from several sources (US Naval
Oceanographic Office, 1958; Weingartner et al.,
1998b) suggest easterly flow is generally 10–20 and
50 cm/s at a maximum. Westerly flow velocities
appear to be similar in magnitude (Munchow et al.,
1998; Weingartner et al., 1998b). Of importance
for sediment dispersal is the observation that in the
Laptev and East Siberian Seas the relative strength
of offshore versus alongshore flow varies. However, this interplay between coastal and offshore
flow is poorly characterized, as the amplitude of
these oscillations in current strength is uncertain
and their frequency unknown.
Ice rafting is another important sediment transport mechanism that occurs primarily during fall
freeze-up. There are two main types of ice: fast ice,
which is attached to shore and immobile in winter
(Timokhov, 1994), and sea ice, drifting ice found
seaward of the fast ice. While sea ice is typically
compacted against the northern fast ice edge in the
Chukchi and East Siberian Seas, it frequently
moves offshore in the Laptev Sea creating a
polynya in which new sea ice continuously forms
and is advected seaward (Barnett, 1991). Conditions in the polynya, i.e. open, shallow and
sometimes turbulent water, are primed for finegrained bottom sediments to be resuspended,
entrained by sea ice and exported from the shelf
ARTICLE IN PRESS
1206
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
(Pfirman et al., 1990; Dethleff et al., 1993;
Nurnberg
.
et al., 1994; Reimnitz et al., 1994;
Eicken et al., 1997). These ice rafted sediments
may be transported to the central Arctic, Canadian shelf, or even North Atlantic before being redeposited as the sea ice melts (Pfirman et al., 1990;
Meese et al., 1997).
3. Methods
In a previous study (Viscosi-Shirley et al., 2003)
we identified several geochemical endmembers in
Siberian shelf surface sediments, each endmember
representing lithogenic sediment input from a
different source region. Here we determine the
relative magnitude of sediment transport along
various pathways on the Siberian shelf based on
spatial gradients in sedimentary geochemical-endmember flux distributions. Using geochemical data
generated in Viscosi-Shirley et al. (2003) for 81
surface sediment samples, we first quantify the
percent abundance of each endmember in each
surface sediment sample by constrained leastsquares regression. (See Fig. 1b for a map of
sample locations and approximate sample depths,
and Viscosi-Shirley et al. (2003) Appendix B for
sample latitudes/longitudes and geochemical results.) We then compile published data for the
Chukchi and Laptev Seas characterizing mass
accumulation rates and linear sedimentation rates,
which we subsequently convert to mass accumulation rates. Endmember fluxes are calculated by
combining our estimated endmember abundances
with the published sediment accumulation rates.
Note that sediment samples in the Chukchi Sea are
generally from within the 0–4 cm depth interval,
and those in the Laptev Sea are typically from
within the top 0–3 cm. Given observed sedimentation rates (discussed below) and surface mixed
layer depths, each sample consequently represents
average sediment transport conditions over
roughly the past century to millenium of sedimentation. The following sections specify the techniques and strategies used in the (1) surface-sediment
sample partitioning; (2) sedimentation rate data
compilation; and (3) sedimentary geochemicalendmember flux calculations.
3.1. Partitioning surface sediments according to
their sources
We partition the chemical composition of the
surface sediment samples to determine the contribution from each endmember, or sediment
source, using constrained least-squares regression
(Menke, 1984). This partitioning of surface sediment composition is based on eight chemical
variables (Si, Al, K, Mg, Sr, La, Ce, Nd) and four
endmember compositions (shale, basalt, mature
sandstone and immature sandstone) (Table 1). As
described in detail in Viscosi-Shirley et al. (2003),
endmember compositions were defined by assuming compositionally extreme samples within the
geochemical data set are representative of the
endmembers. Thus, for each sample we have eight
equations in four unknowns (unknowns in italics):
½Sisample
¼ ð½Sishale %shale Þ þ ð½Sibasalt %basalt Þ
þ ð½Simature
sandstone
þ ð½Siimmature
%mature
sandstone
sandstone Þ
%immature
½A1sample ¼ ð½A1shale %shale Þ þ ? :
sandstone Þ;
ð1Þ
ð2Þ
We solve these equations with the constraints that
(1) endmember percent contributions are greater
than or equal to zero, and (2) the sum of squared
residuals is minimized. Since this set of equations
is solved for each of the 81 samples, if we let n be
the number of samples, m the number of variables
and p the number of endmembers, we can write the
full set of equations in matrix form as:
Enm ¼ Anp Dpm ;
where E is the data matrix, D is the endmember
composition matrix, and A is the endmember
contribution matrix that we are solving for. In
determining a solution, if the data are unweighted,
more abundant elements are fit more closely than
less abundant elements. Thus, we weight the data
such that the weights are inversely proportional to
the variable values in each sample, as specified in
Table 1. Note that while the raw data span over 50
orders of magnitude, the scaled values only vary
by a factor of four.
ARTICLE IN PRESS
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
1207
Table 2
Siberian shelf sedimentation rates
Chukchi
Sea
Source
Core no.
Latitude
Longitude
Baskaran and
Naidu (1995)
SU-5
CH-40
CH-38
CH-21
USGS84-12
CH-25
CH-13
67.03
70.28
70.70
71.20
71.48
72.63
72.52
169.00
167.91
167.38
164.20
165.12
167.08
164.13
Huh p. comm.
NW63-17
USGS85
28-38
USGS85
29-41
USGS85
30-42
USGS85
31-43
67.47
71.60
71.60
71.64
71.51
170.37
165.81
166.26
167.08
167.68
Kulikov et al.
(1970)
Laptev
Sea
South of Wrangel Isl.
Mass
accumulation
rate
(mg/cm2/yr)
Comments
159
119
156
115
324
83
167
210
Pb
Pb. Baskaran and Naidu determined
MARs in two additional cores, not
included at left due to the evidence
of sediment mixing below the surface
mixed layer in the portion of the
210
Pb profile, used to estimate
accumulation rates.
196
281
328
697
300
223
320
374
794
342
210
0–50
0–57
(29-see
comment)
MARs in this region are much lower
than elsewhere on the shelf, ranging from
0 mg/cm2/yr to only 57 mg/cm2/yr, and
we take the midpoint of this range as
typical of MARs in this area.
Linear
sedimentation
rate
(cm/kyr)
210
Pb
Bauch et al.
(2001b)
PM9499
KD9502
PS51/135-4
PS51/141-2
PM9462
PS51/092-13
PS51/092-12
PS51/080-13
75.50
76.20
76.20
75.00
74.50
74.45
74.45
73.50
115.55
133.12
133.20
128.50
136.00
130.25
130.10
131.50
13
3
2
52
52
70
41
28
11
3
2
45
45
61
36
24
14
C (via accelerator mass spectrometry).
The estimated sedimentation rates are
for time intervals o0–5.3 kyr, except
for in cores KD9502 and PS51/092-12,
where the sedimentation rates are for
the time interval B0–7 kyr.
Johnson-Pyrtle
(1999)
24
71.63
128.97
59
52
210
Pb and
137
Cs
Note: Given linear sedimentation rate data, we estimate mass accumulation rates assuming an average grain density of 2650 mg/cm3
and porosities of 57% and 67% in the Chukchi and Laptev Seas, respectively.
3.2. Compilation of sedimentation rate data
In developing a database characterizing modern
Siberian-shelf mass accumulation rates, we use a
combination of 210Pb and 14C based estimates of
sedimentation rates. Because published mass
accumulation rate data (mg/cm2/yr) are scarce,
we also employ published linear sedimentation
rate data (cm/yr) in our calculations. The linear
sedimentation rates are converted to mass accumulation rates as described in Appendix A. (This
calculation was made for entries in Table 2 for
which both linear sedimentation rates and mass
accumulation rates are listed.) It is clearly applicable to include sedimentation rates determined by
the 210Pb technique, which is used for estimating
rates of deposition over the past 100 years (Faure,
1986). While 14C is used to establish sedimentation
rates over longer timescales (45,000 years) (Bard,
1998), for our study carefully selected 14C
based estimates can be employed in conjunction
with the 210Pb based estimates. During the last
ARTICLE IN PRESS
1208
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
deglaciation, rising sea level flooded most of the
Chukchi shelf by 12.8 calendar kyr B.P. (date
originally from Elias et al., 1996; here has been
converted from 14C to calendar years based on the
polynomial equation given in Bard, 1998), central
Laptev shelf by B7 calendar kyr B.P. and outer
Laptev shelf by B9 calendar kyr B.P. (Bauch et al.,
2001a,b). Bauch et al. find that in the Laptev Sea
14
C-based sedimentation rates have been fairly
constant since the transgression was complete,
in particular over the past 4 kyr. In the
Chukchi Sea, there is no evidence of lithologic
changes in post transgressive sediments (Elias et al.,
1996), suggesting that during this period Chukchi
sedimentation rates have been relatively stable as
well. Thus we limit ourselves to 14C based
sedimentation rates derived for post transgressive
sediments that will likely yield values reasonably
comparable to those established by the 210Pb
technique.
Several criteria were established for the selection
of sedimentation rate data. We use only 14C-based
sedimentation rates calculated from 14C ages that
were corrected for fractionation and reservoir
effects and converted to calendar years. In the
publications chosen as sources of 210Pb-based
sedimentation rates, the effects of sediment mixing
on downcore 210Pb profiles have been constrained
in order to accurately quantify sediment accumulation rates from the 210Pb data. Note that 210Pb
based sedimentation rates are derived from
the slope of the linear regression of downcore
210
Pbexcess data plotted versus cumulative dry
mass, and for the cores used in our endmember
flux estimations these regressions have R’s>0.816
(n=4–11) in Baskaran and Naidu (1995) and an R
of 0.941 (n ¼ 5) in Huh’s work (personal communication).
3.3. Estimating endmember fluxes
We determine the flux of material from each
source, or endmember, to each sample location as:
fluxshale ðfor exampleÞ ¼ % abundanceshale
MARbiogenic-free ;
where % abundanceshale is the contribution the
shale endmember makes to a given sample, as
calculated in our partitioning model, and
MARbiogenic-free is the biogenic-free mass accumulation rate at that sample location, estimated from
published data. Modeled endmember percent
abundances describe endmember concentrations
within the sedimentary lithogenic fraction since the
geochemical data used to partition sediment
composition are expressed on an organic free
basis. Thus, we must use lithogenic fraction, or
biogenic-free, mass accumulation rates (MARs) to
estimate endmember fluxes. Biogenic-free MARs
are estimated as:
MARbiogenic-free ¼ MARtotal ð% biogenic matter MARtotal Þ;
where:
% biogenic matter ¼ % biogenic opal þ % carbonate
þð2:5 % organic carbonÞ:
Organic carbon contents are multiplied by 2.5 to
estimate sedimentary organic matter concentrations. Sedimentary biogenic opal, carbonate and
organic carbon data for the surface sediment
samples were provided by Mammone and are
discussed elsewhere (Mammone, 1998). Total
MAR data are compiled from published studies,
as described above. An error analysis for
the calculated endmember fluxes is included in
Appendix B.
4. Results
Modeled percent abundances of the geochemical
endmembers in Siberian shelf sediments are shown
in Fig. 3. To evaluate the model’s effectiveness at
partitioning the sediment samples into contributions from the endmembers consider the coefficients of determination listed in Table 1, and
Fig. 4, which compares measured element concentrations with those predicted by the leastsquares regression model. The coefficients of
determination express for each variable the percent
variance in the data explained by the model. The
model does a good job predicting sample Sr and
rare earth element contents (coefficients of determination greater than 91%). Si, Al and Mg are
predicted fairly well (coefficients of determination
greater than 81%), while K is predicted relatively
ARTICLE IN PRESS
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
1209
Fig. 3. Modeled geochemical endmember % abundances in Siberian shelf sediments. Black lines in the ‘‘Shale’’ panel indicate the
location of the Lena and Yana Rivers’ submarine channels.
poorly (coefficient of determination of 66%). The
endmembers were originally identified, in part, on
the basis of factor analysis of the geochemical data
(Viscosi-Shirley et al., 2003). Endmember percent
abundance distributions seen here are similar to
previously observed factor distributions. This
correspondence between factor and endmember
distributions and the fairly high coefficients of
determination suggest we have done a reasonable
job defining the primary geochemical endmember
compositions present in Siberian shelf surface
sediments and partitioning the surface sediment
samples according to these endmembers.
The compiled published sedimentation rate data
and their sources are summarized in Table 2.
While sediment accumulation rate data are available for Chukchi and Laptev sediments, there are
no published sedimentation rates for East Siberian
sediments. Thus we focus on the Chukchi and
Laptev Seas in this investigation. The distributions
of Chukchi and Laptev shelf sedimentation rates
are shown respectively in Figs. 5a and b. Note that
Chukchi shelf MARs generally exceed those on the
Laptev shelf.
Estimated sedimentary geochemical-endmember
fluxes, their standard deviations, and the data used
in the flux calculations are summarized in Table 3.
With the exception of region B (see below),
endmember fluxes are calculated at locations
specified by the distribution of available MAR
data. In the Chukchi Sea, these locations include
region A (inner central shelf), region B (midcentral shelf), and region C (western shelf
near Wrangel Island) (Fig. 5a). In the Laptev
Sea, endmember fluxes are determined at
individual core locations D through L, which
similarly span north/south and east/west trending
gradients (Fig. 5b). In region B, there are no
measured total MARs and we infer accumulation
rates based on gradients in existing central- and
ARTICLE IN PRESS
40
20
0
0
20
40
Al wt% predicted
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
Si wt% predicted
1210
10
5
0
0
5
Al wt% measured
4
2
0
0
2
4
Mg wt% predicted
K wt% predicted
Si wt% measured
2
1
0
0
300
200
100
0
100
200
300
40
20
0
0
0
0
50
100
Ce ppm measured
20
40
60
La ppm measured
Nd ppm predicted
Ce ppm predicted
50
2
60
Sr ppm measured
100
1
Mg wt% measured
La ppm predicted
Sr ppm predicted
K wt% measured
0
10
40
20
0
0
20
40
Nd ppm measured
Fig. 4. Siberian-shelf surface-sediment measured element concentrations plotted versus values predicted by the least-squares regression
model, which describes sample compositions as mixtures of four geochemical endmembers.
eastern-Chukchi-shelf total MAR data. Centraland eastern-shelf total MARs gradually decrease
offshore as sediment sources become more distant,
from the inner shelf (159–223 mg/cm2/yr), to the
central and outer shelf (115–167 mg/cm2/yr), and
finally approaching the shelf break (83 mg/cm2/yr).
The only exception to this trend is between Herald
and Hanna shoals, where total MARs are at a
maximum (>300 mg/cm2/yr) (Baskaran and Naidu, 1995; Huh, personal communication). Given
the general offshore decline in sedimentation rates,
we assume that region B total MARs are roughly
119–156 mg/cm2/yr, as seen in two cores just
offshore from here. For the Chukchi Sea, we
determine fluxes of the basalt endmember
(Figs. 6a–c) and shale endmember (Figs. 7a–c) in
ARTICLE IN PRESS
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
83
167
Chukchi Sea
Hanna Shl.
794 374 320
324
324
115 -40
Herald Shl. 156
-30
119
Wrangel Isl.
C
-50
-40
-30
70˚
0-57
B
-50
Alaska
Siberia
223
A
159
-160˚
180˚
-170˚
(a)
New Siberian
Ils.
Laptev Sea
L 11
-50
75˚
J3 2 K
I45
G
H 61
36
-20
24
F
1211
For the Laptev Sea, we calculate shale endmember fluxes (Figs. 8a–c) at individual core
locations D–L. To make these calculations, we
grid shale endmember abundances and biogenic
matter concentrations at our sample locations to
estimate values for these parameters at sites D–K.
Site L lies west of the area encompassed by the
surface sediment samples and thus the gridded
data. Here we assume contents equivalent to those
at our westernmost sample location and the one
closest to site L. This is a conservative assumption
since the shale endmember is discharged to the
eastern Laptev Sea and its percent abundance,
which decreases westward with increasing distance
from its source, has already dropped close to zero
in our westernmost sample. We find that shale
endmember fluxes range from o1 to 37 mg/cm2/
yr, and are generally highest close to the Lena
River delta (Fig. 8c).
45
E
-20
-20
5. Discussion
52 D
70˚
Siberia
5.1. Sediment source strength
Indigirka R.
110
˚
Yana R.
Lena R.
120˚
(b)
130˚
140˚
Fig. 5. Total mass accumulation rate data (mg/cm2/yr) for the
(a) Chukchi Sea and (b) Laptev Sea. Data from Baskaran and
Naidu (1995) (squares); Huh (personal communication)
(crosses); Bauch et al. (2001b) (triangles); Johnson-Pyrtle
(1999) (diamond); and Kulikov et al. (1970) (region C). Our
geochemical surface-sediment sample locations are shown for
reference (circles). See text for discussion of regions A–C.
regions A–C. In order to make the flux calculations on a regional basis, we use the average
endmember adundances and biogenic matter concentrations for each region. Basalt endmember
fluxes are 92–129 mg/cm2/yr in region A and a
fairly comparable 88–114 mg/cm2/yr in region B,
whereas they are much lower in region C, B3 mg/
cm2/yr (Fig. 6c). In contrast, shale endmember
fluxes increase to the north/northwest, from
0.0 mg/cm2/yr in region A, to 3–4 mg/cm2/yr in
region B, and 4 mg/cm2/yr in region C (Fig. 7c).
Endmember percent abundances are useful in
evaluating the relative contributions of the endmembers and thus the strength of individual
lithogenic sediment sources on different parts of
the Siberian shelf. A detailed discussion of the
provenance of the sedimentary geochemical endmembers is presented in Viscosi-Shirley et al.
(2003); here, we focus on the implications of
endmember percent abundances for sediment
source strength. Because our samples represent
on the order of several centuries of sedimentation,
findings regarding sediment source strength and
accumulation patterns (see below) reflect the net
effect of variability in precipitation, river outflow,
currents and sediment ice rafting over this time
period. The Laptev Sea receives detrital inputs
from multiple sources. Western Laptev shelf
sediments are 40–80% immature sandstone endmember (Fig. 3). Possible sources of the immature
sandstone endmember include sedimentary rocks
overlying the Siberian platform and New Siberian
Islands. This endmember consists of coarsegrained sediment and may also represent a lag
1212
Table 3
Siberian-shelf sedimentary geochemical endmember fluxes
Laptev Sea
Location D
Location E
Location F
Location G
Location H
Location I
Location J
Location K
Location L
52
24
45
61
36
45
3
2
11
12
12
12
12
7
140
196
105
137
27
6
5
2
5
5
5
4
4
3
49
23
44
58
34
43
3
2
11
Shale
0
0
3
3
13
77
71
35
62
62
48
65
65
7 maximum
Basalt
66
66
83
83
11
Shale
0
0
3
4
4
NAb
NAb
8
9
NAb
37
16
15
36
21
21
2
1
0.8 maximum
NAb
5
9
11
6
11
0.6
0.3
1
Basalt
92
129
88
114
3
25
62
32
18
NAb
a
The average endmember abundances and biogenic matter concentrations for each region are used in order to make flux calculations on a regional basis. Region A
contains only sample and we assume its characteristics are representative of conditions throughout region A. This appears to be reasonable since samples in the general
vicinity of region (south of 69
N and east of 173
W) are similar. For example, they have an average basalt endmember abundance of 68% with a standard deviation of
only 4%.
b
We cannot estimate errors in these fluxes as there is insufficient information regarding accumulation rates or the calculated fluxes are zero.
ARTICLE IN PRESS
Chukchi Seaa
Region A
159
223
Region B
119
156
Region C
29
Endmember fluxes Standard Endmember fluxes Standard
(mg/cm2/yr)
deviation (mg/cm2/yr)
deviation
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
Total mass
Biogenic matter Biogenic-free Endmember % abundances
accumulation rates
(wt%)
MARs
(MARs) (mg/cm2/yr)
(mg/cm2/yr)
ARTICLE IN PRESS
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
-50
-40
Herald Shl.
-30
B
105-137
Alaska
Alaska
-50
66
C
-40
-30
27
-50
Siberia
Hanna Shl.
Wrangel Isl.
-40
-40
1 -30 C
17
Herald Shl.
14
-30
87
13
82 B
100
84
0
83 93 85
95
91 77 68 68
68 45 59
86
8 78
70˚
Chukchi Sea
Hanna Shl.
Wrangel Isl.
-50
Chukchi Sea
1213
Siberia
A
196
140
A
-160˚
180˚
-170˚
(a) Basalt Endmember Percentages
(b) Biogenic-free Mass Accumulation Rates
Chukchi Sea
Hanna Shl.
Wrangel Isl.
C
-50
3
-40
-40
-30
Herald Shl.
-30
B
88-114
-50
Alaska
Siberia
Chukotka
Peninsula
129
92
A
Bering
Strait
(c) Basalt Endmember Fluxes
Fig. 6. (a) Chukchi Sea basalt endmember % abundances; (b) biogenic-free mass accumulation rate data (mg/cm2/yr); and (c) basalt
endmember fluxes (mg/cm2/yr) for regions A–C. Endmember fluxes are calculated by multiplying the regional mass accumulation rate
data in (b) by the average basalt endmember abundance in each region, determined on the basis of data shown in (a) (see text for
details). Symbols as in Fig. 5.
deposit resulting from particle reworking by
currents and ice. The presence of the basalt and
shale endmembers in Laptev Sea sediments can be
clearly linked to inputs from the Siberian flood
basalt, and the Lena and Yana Rivers, respectively. While the basalt endmember accounts for
20% or less of western Laptev shelf sediments, the
shale endmember comprises up to 40% of
sediments in this area. Thus detritus originating
from the eastern Laptev Sea makes up a significant
fraction of western Laptev shelf sediments relative
to local inputs. Eastern Laptev Sea sediments are
likewise a mixture of the shale, basalt and
sandstone endmembers. Here, however, local
inputs predominate with the Lena/Yana Riverderived shale endmember most prevalent. On the
basis of isotopic data, Eisenhauer et al. (1999)
similarly report that Laptev Sea sediments reflect
inputs from the Siberian flood basalt province and
Lena River drainage basin, as well as sediment
reworking to separate coarse and fine grained
material. Their analysis of Nd isotope data, which
ARTICLE IN PRESS
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
Chukchi Sea
-50
C
-40
-40
-30
27
Herald Shl.
-30
B
105-137
Alaska
-50
Alaska
-50
Siberia
Hanna Shl.
Wrangel Isl.
-40
-40
12 -30 C
10
Herald Shl.
7
-30
22
0
9
0 B
0 17 3
7
5
5
9
4 0 0
0 00 0 0
8
0
70˚
Chukchi Sea
Hanna Shl.
Wrangel Isl.
-50
1214
Siberia
A
196
140
A
-160˚
180˚
-170˚
(a) Shale Endmember Percentages
(b) Biogenic-free Mass Accumulation Rates
Chukchi Sea
Hanna Shl.
Wrangel Isl.
C
-40
-50
-40
-30
Long
Strait 4
Herald Shl.
-30
3-4
B
-50
Alaska
Siberia
0
0
A
(c) Shale Endmember Fluxes
Fig. 7. (a) Chukchi Sea shale endmember % abundances; (b) biogenic-free mass accumulation rate data (mg/cm2/yr); and (c) shale
endmember fluxes (mg/cm2/yr) for regions A–C. Endmember fluxes are calculated by multiplying the regional mass accumulation rate
data in (b) by the average shale endmember abundance in each region, determined on the basis of data shown in (a) (see text for
details). Symbols as in Fig. 5.
are not affected by grain size sorting, indicates
the mean isotopic values of western Laptev
samples lie between mean values for the Siberian
flood basalt and eastern Laptev Sea. From these
data they estimate that only about 25% of western
Laptev sediments are derived from the Siberian
flood basalt province. In the Chukchi Sea, sediments are more homogeneous in composition.
Basalt endmember rich sediments originating from
the Okhotsk-Chukotsk volcanic belt and Bering
Strait inflow dominate, constituting over 60% of
Chukchi shelf sediments except near Wrangel
Island and along sections of the Siberian coast.
Sediments in these areas mainly reflect local input
from the Chukotka terrain and erosional processes
(mature sandstone endmember), with some contribution from the East Siberian Sea (shale endmember).
ARTICLE IN PRESS
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
Laptev Sea
0
-20
30
43 66
I43
34 G
H58
-20
F
44
23 E
-20
-20
-20
49 D
86
Siberia
New Siberian
Ils.
J3 2 K
Laptev Sea
L11
39
73
36
33
38 61 20
56 2 8
0 50 67
27 30
55
68
61
66 46
100 19 68
59 30
-20
70˚
New Siberian
Ils.
62
-50
-50
75˚
1215
Siberia
Indigirka R.
110
˚
120˚
(a) Shale Endmember Percentages
Indigirka R.
Yana R.
Lena R.
130˚
Lena R.
Yana R.
140˚
(b) Biogenic-free Mass Accumulation Rates
New Siberian
Ils.
J2 1 K
Laptev Sea
-50
L<0.8
I 21
21 G
H36
-20
F
15
16 E
-20
-20
37 D
Siberia
Indigirka R.
Lena R.
Yana R.
(c) Shale Endmember Fluxes
Fig. 8. (a) Laptev Sea shale endmember % abundances; (b) biogenic-free mass accumulation rate data (mg/cm2/yr); and (c) shale
endmember fluxes (mg/cm2/yr). Endmember fluxes are calculated by multiplying the mass accumulation rate data in (b) by endmember
% abundances at these locations, estimated by gridding the % abundance data in (a) (see text for details). Symbols as in Fig. 5.
5.2. Dominant sediment transport pathways
A key advantage to mapping endmember fluxes,
as opposed to endmember abundances, is that
each endmember’s distribution is unaffected by
dilution from other sources. This allows us to infer
the dominant dispersal pathways for sediment
accumulating on the shelf, based on observed
spatial variability in endmember fluxes. In the
Chukchi Sea portion of our study area, the
distribution of basalt endmember fluxes enables
us to identify its primary transport pathways from
its southern Chukchi source region. Derived from
southern Alaskan volcanic rocks and Siberia’s
Okhotsk-Chukotsk volcanic belt, the basalt endmember is delivered to the Chukchi Sea primarily
via Bering Strait inflow and runoff from the
Okhotsk-Chukotsk volcanic belt (Fig. 6c) (Viscosi-Shirley et al., 2003). Basalt endmember fluxes
are 92–129 mg/cm2/yr in region A, which is located
closest to the basalt endmember’s sources. Basalt
endmember fluxes are almost as high to the north
in region B (88–114 mg/cm2/yr) as they are in
region A, indicating northerly transport of the
basalt endmember is strong. With regard to
northwesterly transport, regions C and B are
comparable distances northwest and north, respectively, of the basalt endmember’s source
region. Yet basalt endmember fluxes in region C
(3 mg/cm2/yr) are roughly one thirtieth of those
in region B, implying northwesterly transport of
the basalt endmember is very weak compared to
ARTICLE IN PRESS
1216
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
northerly transport. The errors in calculated basalt
endmember fluxes for regions A and B are
significant in some cases, ranging from 16% to
48% (Table 3). The error in region C’s basalt
endmember flux is unknown, and we conservatively assume this flux has an error of 48%
(equivalent to the largest estimated error for a
basalt endmember flux). Even given these large
errors, our observations hold that basalt endmember fluxes in regions A and B are fairly
comparable while those in region C are much
lower, as do our conclusions that offshore-northerly transport of this endmember is significantly
stronger than northwesterly transport.
East Siberian Sea sediments, represented by the
shale endmember, enter the Chukchi Sea through
Long Strait (Fig. 7c). In the vicinity of the strait
(region C), shale endmember fluxes are 4 mg/cm2/
yr. East of the strait in the central Chukchi Sea
(region B) shale endmember fluxes are a comparable 3–4 mg/cm2/yr, yet there is no accumulation
of this endmember southeast of the strait on the
inner Chukchi Shelf (region A). These observations suggest the dominant sediment dispersal
pathways for sediments in our Chukchi study area
are from the Bering Strait region to the north and
from Long Strait to the east, into the central
Chukchi Sea. At our Chukchi Sea sample sites
(water depths >40 m), there appears to be
comparatively minimal movement of these sediments parallel to the coast, either to the northwest
or southeast.
The Laptev Sea exhibits a different dominant
sediment dispersal pattern than the Chukchi Sea.
The Lena and Yana Rivers are sources of the shale
endmember in the Laptev Sea (Fig. 8c). Discharging about five times more sediment than the Yana
River (Gordeev et al., 1996; Lisitzin, 1996), the
Lena River is the shale endmember’s primary
source. The Lena delta’s eastern branches supply
roughly 90% of the river’s suspended matter,
with most (64%) of its particulate discharge
coming from a single northeast branch (Kuptsov
and Lisitsin, 1996). Shale endmember fluxes
are highest at sites within several hundred kilometers of the Lena River delta’s main branches
(D-37 mg/cm2/yr [error unknown]; E-1675 mg/
cm2/yr; G-36711 mg/cm2/yr; H-2176 mg/cm2/yr;
I-21711 mg/cm2/yr) and significantly lower elsewhere on the shelf. This result is consistent with
Stein and Fahl’s (2000) observation that fluxes of
organic carbon, which is primarily terrigenous in
origin in the Laptev Sea (Rachold and Hubberten,
1999; Stein and Fahl, 2000), peak off the Lena
River delta and progressively decline offshore.
Shale endmember fluxes are only 170.3 and
270.6 mg/cm2/yr north of the delta at central
shelf locations K and J, or 4–67 times less than
those adjacent to the delta given the errors in the
calculated fluxes. A comparable distance west/
northwest of the Lena River delta at site L, shale
endmember fluxes are o0.871 mg/cm2/yr, or 5–59
times lower than those near the delta. Site F lies
east/northeast of the delta, slightly closer to it than
the northerly (K, J) and westerly (L) sites just
described. Here the shale endmember’s accumulation rate (1579 mg/cm2/yr) is less than typical
fluxes close to the delta but noticeably higher than
at sites K, J and L. This pattern indicates that in
contrast to conditions in the Chukchi Sea, at our
Laptev Sea sample sites (typically >20 m water
depth) sediments originating from the Lena and
Yana Rivers preferentially move to the east/
northeast parallel to the coast, with relatively
minimal transport northward to the mid-shelf.
Similar to the Chukchi Sea, there is comparatively
little dispersal of these sediments to the west/
northwest.
Comparing these results for the Laptev Sea with
observations based on geographic gradients in
shale endmember percentages illustrates the importance of determining dominant sediment transport pathways based on endmember fluxes rather
than endmember percent abundances. Shale endmember percentages are high close to its sources,
just off the mouths of the Lena and Yana Rivers
(Fig. 3). The percentages remain elevated along the
length of these rivers’ northerly trending submarine channels but drop significantly immediately
east and west of the channels, suggesting that
offshore dispersal of these sediments is greater
than that parallel to the coast. As described above,
an examination of shale endmember fluxes provides a different picture. Shale endmember fluxes
are relatively high off the Lena River delta but
decrease sharply to the north while declining
ARTICLE IN PRESS
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
gradually to the east/northeast (Fig. 8c). Thus
shale endmember fluxes indicate that Lena River
detritus accumulating on the shelf tends to move
along the shore to the east/northeast, rather than
offshore to the north as suggested by trends in
percent abundances.
5.3. Controls on sediment dispersal patterns
The spatial and temporal variability in Siberian
shelf currents and ice rafting prevent us from
predicting dominant sediment transport pathways
a priori. However, assuming the variability documented in shelf physical oceanography over
roughly the last half century is typical of that
during the past several centuries represented by
our samples, we can use our physical oceanographic knowledge to get some idea of the factors
controlling observed sediment dispersal patterns.
In the Chukchi portion of our study area, the
dominant sediment transport pathways reflect
currents and bathymetry. Water entering from
Bering Strait tends to flow north at all depths
following bottom contours, then pass around
Herald shoal to exit the region via Hope Sea
Valley and Barrow Canyon rather than moving
northwest into Long Strait (Fig. 2) (Coachman
et al., 1975; Coachman and Shigaev, 1992;
Weingartner et al., 1998a). Consistent with this
observation, Chukchi sedimentary endmember
flux distributions indicate offshore/northerly sediment transport is significantly greater than sediment movement to the northwest, parallel to the
Siberian coast. Surface and bottom inflow from
Long Strait moves east and southeast, at some
point turning north to join the cross shelf flow.
Our results indicate that sediments from Long
Strait accumulate on the central shelf but are
absent from the southern shelf, suggesting the
inflowing waters typically turn north before reaching the innermost Chukchi shelf. The sediments
considered here, those rich in the shale and basalt
endmembers, are likely transported as suspended
load. Both the shale and basalt endmembers are
fine grained, and according to McManus et al.
(1969) Chukchi shelf current velocities are sufficient to transport the silt and clay sized particles
found on the shelf in suspension. Because water
1217
circulation in the study area is fairly coherent at all
depths, it is difficult to distinguish the influence of
surface versus bottom currents on sediment
dispersal patterns. McManus et al. (1969) describe
a turbidity maximum in Chukchi shelf bottom
waters and infer bottom currents are a primary
dispersal mechanism on the shelf.
A combination of river outflow and water
circulation is important in determining sedimentation patterns on the Laptev shelf. River discharge
of water and sediment into the Laptev Sea occurs
mainly in the summer months, with 90% of the
outflow supplied in the warm season (Pavlov et al.,
1996; Behrends et al., 1999). A plume of river
water is evident off the Lena River delta, extending
to the north, northeast, or east depending on the
hydro-meteorological conditions (Pavlov et al.,
1996). The distribution of river outflow corresponds spatially to that of the highest shale
endmember fluxes indicating river discharge is
critical in dispersing the shale endmember. Though
wintertime currents in the Laptev Sea are poorly
characterized, summertime currents are known to
flow in cyclonic gyres both at the surface (Pavlov
et al., 1996) and at depth (Timokhov, 1994; Hass
et al., 1995). Near shore, the Siberian Coastal
Current generally flows to the east with occasional
reversals to the west (Pavlov, 1995; Weingartner
et al., 1998b; Munchow et al., 1998). The northerly
and easterly components of this flow pattern also
undoubtedly contribute to the elevated shale
endmember fluxes off the Lena River delta. It is
important to note that there are fluctuations in the
relative intensity of the northerly versus easterly
branches of cyclonic flow, though the amplitude
and frequency of these oscillations is unknown.
Above we observed that with increasing distance
from the Lena River delta shale endmember fluxes
decline relatively gradually to the east/northeast
but sharply to the north, indicating sediment
dispersal is stronger to the east/northeast than to
the north. One possible explanation for this
sedimentation pattern is that on average during
the time represented by our samples easterly flow
has been strong compared with northerly flow.
Sediment ice rafting is another potential cause
of the low shale endmember fluxes evident on the
northern and western Laptev shelf. Although the
ARTICLE IN PRESS
1218
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
magnitude of sediment entrainment and export
vary a great deal spatially and temporally,
published studies suggest a significant amount of
ice rafted debris is exported from the Laptev shelf
to the central Arctic (Nurnberg
.
et al., 1994;
Dethleff, 1995; Eicken et al., 1997, 2000; Pfirman
et al., 1997). Pfirman et al. (1997) identify the
Laptev’s New Siberian Island region as a source of
.
sediment-laden ice. Holemann
et al. (1999) documented sediment entrainment off the Lena River
delta, in the New Siberian Island region, and in the
western Laptev Sea both near the Khatanga River
mouth and along the coast. Existing budgets place
ice rafted debris export from sections of the central
and eastern Laptev at about 3.5–4 million tons/yr
(Dethleff, 1995; Eicken et al., 1997) and from the
New Siberian Island region at 18.5 million tons/yr
(Eicken et al., 2000). These observations suggest
sediment ice rafting may help cause the low shale
endmember fluxes on the northern and western
Laptev shelf through a combination of factors.
First, much of the inner shelf sediment that is
entrained in sea ice and carried offshore may
bypass the central shelf (sites H and K) and be redeposited farther offshore or exported into the
central Arctic. In addition, our northerly and
westerly sites (H, K and I) are located in zones of
active turbid sea ice formation. Sea ice entrainment and removal of sediment from sites H, K and
I may be significant enough compared with the
magnitude of sediment inputs to these areas to
produce some of the lowest terrigenous fluxes seen
in our study area. Additional data are needed to
test these ideas, including better constrained
estimates of Siberian shelf sediment inputs (via
river discharge, coastal erosion, Bering Strait
inflow, etc.) and outputs (by currents, ice rating,
etc.). Such data also have broader implications for
understanding general sediment accumulation
patterns on the shelf. Our results indicate that
endmember fluxes and biogenic-free MARs in the
Chukchi Sea exceed those in the Laptev Sea (Fig.
5), implying there is greater input relative to
output of lithogenic sediment in the Chukchi Sea.
Such data will help determine why this pattern
exists, when the Laptev Sea receives high sediment
input from rivers (20–25 106 tons/yr; Gordeev
et al., 1996; Lisitzin, 1996; Are, 1999) and coastal
erosion (58.4 106 tons/yr; Rachold et al., 2000),
while the Chukchi Sea lacks large delta
forming rivers and likely receives less river born
sediment.
5.4. Implications
As the globe and the Arctic become increasingly
industrialized, it is critical to consider how
contaminants produced by this industrialization
will impact the environment for future generations. The inferred sediment transport patterns
have important implications for the fate of particle
reactive contaminants released to the Siberian
shelf. A primary concern is the exposure of
biological organisms, and through the food chain
potentially humans, to particle reactive pollutants
that may remain in Arctic ecosystems for a long
time (AMAP, 2002). Compared with the slope and
basin, Arctic continental shelves are sites of
concentrated biological activity. If particle reactive
contaminants were released to the Chukchi Sea,
the comparatively weak sediment transport we
observe parallel to the Siberian coast (water depths
>40 m) would help limit the pollutants contact
with, and thus detrimental effects on, shelf biota.
There is considerable sediment movement from the
inner to the central Chukchi shelf, as evidenced by
the negligible drop in basalt endmember fluxes
from just north of Bering Strait to the mid-shelf.
However, in a core from the Chukchi shelf edge,
Darby et al. (1999, 2002) find that over the last
millenium easterly and westerly sources of Fe
oxide grains are much more important than
southern Chukchi sources, with o5% of Fe oxide
grains originating in the south. The core has a
sedimentation rate of 20 cm/kyr in the middle to
late Holocene. Assuming 5% of bulk surface
sediments at this outer shelf location are derived
from the southern Chukchi shelf implies a flux of
0.8 mg/cm2/yr (for porosity B0.7 and grain
density 2650 mg/cm3), or up to 160 times less than
basalt endmember fluxes on the shelf. Overall,
these observations suggest that much of the
sediment supplied to the Chukchi Sea is sequestered there rather than being transported along the
shore or off the shelf, making this region an
effective trap for particle reactive pollutants.
ARTICLE IN PRESS
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
In contrast, Laptev shelf sediments move
primarily eastward/northeastward, parallel to the
coast. According to Pavlov et al. (1996) there is
significant water flow from the Laptev Sea to the
East Siberian Sea (0.1 Sv/yr on average). As a
result, Laptev sediments may be dispersed eastward into the East Siberian Sea. Additionally, sea
ice exports considerable amounts of ice rafted
debris from the Laptev shelf to the central Arctic
(Nurnberg
.
et al., 1994; Dethleff, 1995; Eicken
et al., 1997, 2000; Pfirman et al., 1997). Behrends
et al. (1999) and Schoster et al. (2000) identified
the Laptev Sea as a source of central Arctic
seafloor sediments by comparing shelf and basin
sediment chemistry and mineralogy. Thus instead
of acting as a sink for particle reactive pollutants
as the Chukchi Sea does, the dominant sediment
dispersal pattern and occurrence of sediment ice
rafting in the Laptev Sea make this region is a
potential pollutant source for both the Siberian
shelf and Arctic basin. In agreement with this idea,
Meese et al. (1997) investigated 137Cs concentrations in ice and ice born sediments collected from
across the Arctic Basin. The sediment sample with
the highest 137Cs content was found in the
Beaufort Sea and had a significantly higher 137Cs
level than bottom sediments in the same area. On
the basis of discriminant function analysis and
ice transport model results, the authors inferred
the Laptev/Kara Sea region to be the source of the
contaminated sediment.
6. Conclusions
We previously demonstrated in Viscosi-Shirley
et al. (2003) that Siberian shelf surface sediments
have distinct, endmember geochemical compositions, representing lithogenic sediment inputs from
different source regions. Here we partition 81 shelf
surface sediment samples (Fig. 1b) to determine
the percent abundances of these endmembers, or
sediment sources, in each sample. These results are
used to evaluate the relative strength of individual
sediment sources at various locations on the shelf.
We then compile published sedimentation rate
data for the Chukchi and Laptev Seas. Sedimentary geochemical endmember fluxes are deter-
1219
mined by combining our calculated endmember
percentages with the published sedimentation rate
data. From spatial variability in sedimentary
geochemical endmember fluxes, we infer the
dominant transport pathways for sediments accumulating in the Chukchi and Laptev portions of
our study area. Comparing the inferred sediment
transport and accumulation patterns with physical
oceanographic data, we look for evidence of the
factors controlling these patterns.
We find that in the western Laptev Sea,
sediments from the Siberian flood basalt (basalt
endmember) account for o20% of surface sediments, while detritus from the Lena and Yana
Rivers (shale endmember) constitutes as much as
40% of seafloor sediments (Fig. 3). The remaining
coarse-grained sediment (immature sandstone endmember) is either eroded from sedimentary rocks
overlying the Siberian platform and New Siberian
Islands, or results from grain size sorting. Thus
sediment from eastern Laptev sources makes up a
significant fraction of western Laptev shelf sediments relative to local inputs. These results are
similar to Eisenhauer et al.’s (1999) findings based
on isotopic analyses that the Siberian flood basalt
province supplies only about 25% of western
Laptev sediments. Sediments in the eastern Laptev
Sea are likewise a mix of the basalt, shale and
immature-sandstone endmembers. Here, however,
local inputs from the Lena and Yana Rivers are
most prevalent. By comparison, Chukchi Sea
sediments are much more homogeneous in composition. Detritus supplied by Bering Strait inflow
and the Okhotsk-Chukotsk volcanic belt makes up
over 60% of Chukchi shelf sediments, except near
Wrangel Island and along portions of the Siberian
coast. Sediments here reflect local input from the
Chukotka terrain and erosional processes, with
some contribution from the East Siberian Sea.
The dominant transport pathways for sediments
in the Chukchi portion of our study area are from
the Bering Strait region to the north and from
Long Strait to the east, into the central Chukchi
Sea (Figs. 6c and 7c). There appears to be
comparatively minimal sediment movement parallel to the Siberian coast at our sample sites (>40 m
water depth), either to the northwest or southeast.
The Laptev Sea exhibits a different dominant
ARTICLE IN PRESS
1220
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
sediment dispersal pattern (Fig. 8c). At our Laptev
Sea sample sites (typically >20 m water depth),
sediments supplied by the Lena and Yana Rivers
move primarily eastward/northeastward parallel
to the coast, with relatively minimal transport
northward to the central shelf. Similar to the
Chukchi Sea, there is comparatively little movement of these sediments to the west/northwest.
Good correspondence between Chukchi-shelf
sediment accumulation patterns and currents,
which are fairly coherent at all water depths,
indicates water circulation is an important sediment transport mechanism in this marginal sea.
Consistent with this observation, sedimentological
analyses by McManus et al. (1969) provide
evidence that the fine sediments predominant on
the Chukchi shelf are transported as suspended
load by bottom currents. A combination of river
outflow, cyclonic water circulation and the Siberian Coastal Current controls sediment distribution
in the Laptev Sea. We speculate that ice rafting,
which is known to export significant amounts of
sediment from the Laptev shelf (Nurnberg
.
et al.,
1994; Dethleff, 1995; Eicken et al., 1997, 2000;
Pfirman et al., 1997), may also affect sedimentation patterns through a combination of factors.
Sediment fluxes are comparatively quite low on the
central shelf near the New Siberian Islands and the
western shore. These regions are zones of active
turbid sea ice formation (Pfirman et al., 1997;
.
Holemann
et al., 1999). In these areas sea ice
entrainment and removal of sediment may be
significant enough compared with the magnitude
of sediment inputs to suppress sediment accumulation rates. Additionally, much of the ice rafted
sediment that is entrained on the inner shelf and
carried offshore may bypass these regions of the
central shelf, and instead be re-deposited on the
outer shelf or in the central Arctic.
While previous studies have identified directions
of sediment movement on the Siberian margin
(Naidu et al., 1982; Nurnberg
.
et al., 1994;
Behrends et al., 1999; Eisenhauer et al., 1999;
Rachold, 1999; Rossak et al., 1999; Wahsner et al.,
1999; Schoster et al., 2000), here we have mapped
flux distributions and identified the dominant
dispersal pathways of sediments in the Chukchi
and Laptev Seas originating from several key
source regions. These results have implications
for the impact of particle reactive pollutants
released to the shelf, allowing us to identify
zones most threatened by contamination. This
information is also essential to accurately characterizing links between modern Arctic-shelf depositional patterns and the physical environment,
and thus is a necessary prerequisite to interpreting
the record of paleoenvironmental change preserved in Arctic sediment stratigraphy. To conclude, we have made an initial attempt at
characterizing modern sediment source strength,
transport pathways, and accumulation patterns on
parts of the Siberian shelf, and we feel it is
important to expand on these efforts. In particular, additional Siberian-shelf MAR estimates are
necessary to provide a clearer picture of recent
sedimentation patterns. To better define the
processes controlling these patterns will require
quantification of sediment transport by currents
and refined estimates of ice rafted debris export
from the shelf.
Acknowledgements
We gratefully acknowledge Henning Bauch,
who allowed us access to his 14C data prior to its
publication. Ilia and Josanna Zolotov translated
work by Kulikov et al. (1970) from Russian to
English. Jack Dymond and Ian Walsh thoughtfully critiqued the draft manuscript. Chris Guay
supplied a file that was modified to create Fig. 1a.
This work was funded by the Office of Naval
Research (grant N00014-9410982 and Augmentation Award for Science and Engineering Research
Training N00014-9311170 (N. Pisias)). Additional
support came from a College of Oceanic and
Atmospheric Sciences Dean’s Scholarship and an
Oregon State University Supplemental Oregon
Laurels Graduate Scholarship awarded to C.
Viscosi-Shirley.
Appendix A. Linear sedimentation rate conversion
Linear sedimentation rates are converted
to total mass accumulation rates (MARs) as
ARTICLE IN PRESS
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
follows:
2
total MAR in mg=cm =yr
¼ ðlinear sedimentation rate in cm=krÞ
ðaverage grain density in mg=cm3 Þ
ð1 kyr=1000 yrÞ ð1 fÞ;
where f is downcore average sediment porosity.
This calculation was made for entries in Table 2
for which both linear sedimentation rates and
MARs are listed. We estimate that the average
grain density is typically about 2650 mg/cm3,
based on the abundance of various mineral types
(Silverberg, 1972; Naugler et al., 1974; Stein and
Korolev, 1994) and their respective densities (Pye,
1994). The primary constituents of shelf sediments
all have similar grain densities (quartz, 2650 mg/
cm3; feldspars, B2650 mg/cm3; clay minerals,
B2600 mg/cm3). Heavy minerals common to the
shelf have notably higher grain densities (3200–
3500 mg/cm3), but comprise less than B10% of
shelf sediments (estimate based on data from
Silverberg (1972) and Naugler et al. (1974)).
Changes in total heavy mineral concentration
appear to be the main cause of variability in the
average grain density, although they typically
amount to variations of less than B73%. Thus
in our error analysis (see Appendix B) we take 3%
as the average grain density’s standard deviation.
Regarding sediment porosity, in the Chukchi Sea,
we use data from Baskaran and Naidu (1995) to
determine downcore average sediment porosities
for six cores collected close to those for which we
have measured linear sedimentation rates. The
average downcore-average porosity of these cores
is 57% with a standard deviation of 75%. Thus
we set f equal to 0.57, and for our error analysis
we take 5% as the standard deviation in Chukchi
sediment porosity. In order to parameterize f for
the Laptev Sea we use sediment porosity data for a
central Laptev core from Kassens et al. (1995). We
consider only measurements made in the upper
25 cm, as this is the approximate length of Laptev
Sea cores for which we have linear sedimentation
rate estimates. Over this depth interval, the core
has an average sediment porosity of 67%, or f
equal to 0.67. Since we cannot determine the
standard deviation in Laptev sediment porosity
1221
based on a single core, for the error analysis we
assume it is equivalent to that of Chukchi
sediments (5%).
Appendix B. Error analysis
Endmember fluxes are estimated as the product
of MAR and endmember percent abundance data.
According to Taylor (1982), when several quantities are multiplied together the fractional error of
the product, defined as the standard deviation
(SD) of the product divided by the product, is the
sum of the quantities’ fractional errors. Thus the
fractional error on an estimated endmember flux is
the sum of the fractional errors in the MAR and
percent abundance values:
SDflux =flux ¼ SDMAR =MAR
þ SD%
abundance =%
abundance
and the error on the flux is consequently:
SDflux ¼ ðSDMAR =MAR
þ SD% abundance =% abundanceÞ flux:
Note that although biogenic-free MARs are
used in calculating endmember fluxes, to simplify
our error analysis we approximate the fractional
errors in the biogenic-free MARs as the fractional
errors in the corresponding total MARs. This
approximation is reasonable since the biogenic
matter concentrations used in converting total to
biogenic-free MARs are small values with relatively high accuracies and consequently excluding
the errors in these terms does not noticeably
change the estimated errors in endmember fluxes.
Where linear sedimentation rates are used
instead of MARs, fractional errors for average
sediment porosity and density are included in the
equation. Fractional errors for these terms are
calculated from their standard deviations, specified
in Appendix A.
To determine the fractional errors in endmember percent abundances that result from inaccuracies in measured surface-sediment sample
multi-element chemistry we use a Monte Carlo
approach. Geochemical data for six surface sediment samples, chosen for their diverse compositions, which represent both extreme and typical
ARTICLE IN PRESS
1222
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
shelf sediment chemistries, form the basis of this
part of the error analysis. Given each sample’s
elemental composition and each element’s accuracy, we use a random number generator to create
six data sets whose mean elemental concentrations
equal that of the samples and where each element’s
standard deviation is defined by that element’s
accuracy. We partition the resulting data sets using
the constrained least squares technique described
in the Methods. Taking the standard deviation of
each endmember’s percent abundance in each of
the six data sets yields six estimates of the error on
that endmember’s percent abundance values, given
the accuracy of the geochemical data. Based on the
calculated standard deviations, we find modeled
endmember abundances are accurate to within:
7% for the shale endmember; 10% for the basalt
endmember; 4% for the mature sandstone endmember; and 5% for the immature sandstone
endmember.
The approach for calculating fractional errors
in the MARs varies depending on whether the
sedimentation rates are 210Pb or 14C based
estimates. 210Pb based MARs and linear sedimentation rates are calculated by regressing
Ln210Pbexcess versus cumulative dry mass and
depth, respectively. Fractional errors in these
estimates are equivalent to the fractional errors
on the slopes of the regressions, or SDslope/slope,
where the standard deviation of the slope is
calculated as (Draper and Smith, 1966):
SDslope ðe:g: for a linear sedimention rateÞ
X
¼ sqrt
ðLn210 Pbexcessmean Ln210 Pbexcess Þ2
ð1 R2 Þ=ðn 1Þ
X
Csqrt
ðdepth mean depthÞ2 :
14
C-based sedimentation rates used in the endmember flux calculations are determined based on
a single 14C age A (in kyr) at depth z (in cm) as
z=A: Given the error on the 14C age, E; each
calculated sedimentation rate may range from
z=ðA þ EÞ to z=ðA EÞ: Based on the statistics of
small sample sets (Wetherill, 1967), when two
values describe the range of an estimate the error
on the estimate is equal to the range multiplied by
0.886. Thus, 14C based sedimentation rates have a
fractional error of SDsedimentation rate/sedimentation
rate where SDsedimentation rate is:
SDsedimentation
rate
¼ 0:886 ðz=ðA EÞ z=ðA þ EÞÞ:
Endmember flux errors are summarized in Table 3.
References
AMAP, 2002. Arctic Pollution 2002. AMAP, Oslo, Norway,
pp. 1–112.
Are, F.E., 1999. The role of coastal retreat for sedimentation
in the Laptev Sea. In: Kassens, H., Bauch, H.A.,
Dmitrenko, I., Eicken, H., Hubberten, H.-W., Melles, M.,
Thiede, J., Timokhov, L. (Eds.), Land–Ocean Systems in the
Siberian Arctic: Dynamics and History. Springer, Berlin,
pp. 287–295.
Bard, E., 1998. Geochemical and geophysical implications of
the radiocarbon calibration. Geochimica et Cosmochimica
Acta 62 (12), 2025–2038.
Barnett, D., 1991. Sea ice distribution in the Soviet Arctic. In:
Brigham, L. (Ed.), The Soviet Maritime Arctic. Naval Inst.
Press, Annapolis, MD, pp. 47–62.
Baskaran, M., Naidu, A.S., 1995. 210Pb-derived chronology and
the fluxes of 210Pb and 137Cs isotopes into continental shelf
sediments, East Chukchi Sea, Alaskan Arctic. Geochimica
et Cosmochimica Acta 59, 4435–4448.
Bauch, H.A., Kassens, H., Naidina, O.D., Kunz-Pirrung, M.,
Thiede, J., 2001a. Composition and flux of holocene
sediments on the eastern Laptev Sea shelf, Arctic Siberia.
Quaternary Research 55 (3), 344–351.
Bauch, H.A., Muller-Lupp,
.
T., Taldenkova, E., Spielhagen,
R.F., Kassens, H., Grootes, P.M., Thiede, J., Heinemeier,
J., Petryashov, V.V., 2001b. Chronology of the Holocene
transgression at the North Siberian margin. Global and
Planetary Change 31 (1–4), 123–137.
Behrends, M., Hoops, E., Peregovich, B., 1999. Distribution
patterns of heavy minerals in Siberian Rivers, the Laptev
Sea and the Eastern Arctic Ocean: an approach to identify
sources, transport and pathways of terrigenous matter. In:
Kassens, H., Bauch, H.A., Dmitrenko, I., Eicken, H.,
Hubberten, H.-W., Melles, M., Thiede, J., Timokhov, L.
(Eds.), Land–Ocean Systems in the Siberian Arctic:
Dynamics and History. Springer, Berlin, pp. 265–286.
Coachman, L.K., Shigaev, V.V., 1992. Northern BeringChukchi Sea ecosystem: the physical basis. In: Nagel, P.A.
(Ed.), Results of the Third Joint US–USSR Bering and
Chukchi Seas Expedition (BERPAC), Summer 1988. US
Fish and Wildlife Service, Washington, DC, pp. 17–27.
Coachman, L.K., Aagaard, K., Tripp, R.B., 1975. Bering
Strait, The Regional Physical Oceanography. University of
Washington Press, Seattle, WA, pp. 1–172.
ARTICLE IN PRESS
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
Darby, D.A., Bischof, J.F., Cutter, G.A., Dwyer, G.S., Polyak,
L., de Vernal, A., Hillaire-Marcel, C., McManus, J.,
Osterman, L., Poore, R.Z., 1999. Collaborative research:
paleoceanographic records of western Arctic Shelf–Basin
interactions. Final Report to NSF, Div. of Polar Programs
re OPP-9817051.
Darby, D.A., Bischof, J., deVernal, A., Hillaire-Marcel, C.,
2002. A millennial-scale paleoclimate record from the
western Arctic Ocean. EOS 83, F787.
Dethleff, D., 1995. Sea ice and sediment export from the Laptev
Sea flaw lead during 1991/92 winter season. In: Kassens, H.,
Piepenburg, D., Thiede, J., Timokhov, L., Hubberten,
H.-W., Priamikov, S.M. (Eds.), Berichte zur Polarforschung
176, 78–93.
Dethleff, D., Nurnberg,
.
D., Reimnitz, E., Saarso, M.,
Savchenko, Y.P., 1993. East Siberian Arctic Region
Expedition ’92: the Laptev Sea—its significance for Arctic
sea-ice formation and transpolar sediment flux. In: Reimann, F. (Ed.), Berichte zur Polarforschung 120, 3–44.
Dmitrenko, I.A., TRANSDRIFT II Shipboard Scientific Party,
1995. The distribution of river run-off in the Laptev Sea: the
environmental effect. Russian–German cooperation: Laptev
Sea system. In: Kassens, H., Piepenburg, D., Thiede, J.,
Timokhov, L., Hubberten, H.-W., Priamikov, S.M. (Eds.),
Berichte zur Polarforschung 176, 114–120.
Draper, N.R., Smith, H., 1966. Applied Regression Analysis.
Wiley, New York, pp. 1–407.
Eicken, H., Reimnitz, E., Alexandrov, V., Martin, T., Kassens,
H., Viehoff, T., 1997. Sea-ice processes in the Laptev Sea
and their importance for sediment export. Continental Shelf
Research 17 (2), 205–233.
Eicken, H., Kolatschek, J., Freitag, J., Lindemann, F., Kassens,
H., Dmitrenko, I., 2000. A key source area and constraints
on entrainment for basin-scale sediment transport by Arctic
sea ice. Geophysical Research Letters 27 (13), 1919–1922.
Eisenhauer, A., Meyer, H., Rachold, V., Tutken,
.
T., Wiegand,
B., Hansen, B.T., Spielhagen, R.F., Lindemann, F.,
Kassens, H., 1999. Grain size separation and sediment
mixing in Arctic Ocean sediments: evidence from the
strontium isotope systematic. Chemical Geology 158,
173–188.
Elias, S.A., Short, S.K., Nelson, C.H., Birks, H.H., 1996. Life
and times of the Bering land bridge. Nature 382, 60–63.
Faure, G., 1986. Principles of Isotope Geology. Wiley,
New York, pp. 1–589.
Gordeev, V.V., Martin, J.M., Sidorov, I.S., Sidorova, M.V.,
1996. A reassessment of the Eurasian river input of water,
sediment, major elements, and nutrients to the Arctic
Ocean. American Journal of Sciences 296, 664–691.
Hass, H.C., Antonow, M., Shipboard Scientific Party, 1995.
Movement of Laptev Sea shelf waters during the transdrift
II expedition. Russian–German cooperation: Laptev Sea
system. In: Kassens, H., Piepenburg, D., Thiede, J.,
Timokhov, L., Hubberten, H.-W., Priamikov, S.M. (Eds.),
Berichte zur Polarforschung 176, 121–134.
.
Holemann,
J.A., Schirmacher, M., Prange, A., 1999. Dissolved
and particulate major and trace elements in newly formed
1223
ice from the Laptev Sea (Transdrift III, October 1995). In:
Kassens, H., Bauch, H.A., Dmitrenko, I., Eicken, H.,
Hubberten, H.-W., Melles, M., Thiede, J., Timokhov, L.
(Eds.), Land–Ocean Systems in the Siberian Arctic:
Dynamics and History. Springer, Berlin, pp. 101–111.
Holmes, M.L., 1975. Tectonic framework and geologic evolution of the southern Chukchi Sea continental shelf. Ph.D.
Thesis, University of Washington.
Holmes, M.L., Creager, J.S., 1974. Holocene history of the
Laptev Sea continental shelf. In: Herman, Y. (Ed.), Marine
Geology and Oceanography of the Arctic Seas. Springer,
Berlin, pp. 211–229.
Johnson-Pyrtle, A., 1999. Distribution of 137Cs in the Lena
River Estuary–Laptev Sea system as evidenced by marine,
estuarine and lacustrine sediments. Ph.D. Thesis, Texas
A&M University.
Kassens, H., Thiede, J., 1994. Climatological significance of
arctic sea ice at present and in the past. Russian–German
Cooperation in the Siberian shelf seas: geo-system LaptevSea. In: Kassens, H., Hubberten, H.-W., Pryamikov, S.M.,
Stein, R. (Eds.), Berichte zur Polarforschung 144, 81–85.
.
Kassens, H., Bauch, H., Cremer, H., Dehn, J., Holemann,
J.,
Kunz-Pirrung, M., Peregovich, B., 1995. The depositional
environment of the Laptev Sea. In: Kassens, H. (Ed.),
Berichte zur Polarforschung 182, 86–97.
Kulikov, N.N., Lapina, N.N., Semenov, Y.P., Belov, N.A.,
Spiridonov, M.A., 1970. Stratifikatsiya i skorosti nakopleniya donnykh otlozheniy arkticheskikh morey SSSR. In:
Severnyy Ledovityy okean i yego poberezh’ye v kaynozoye;
Istoriya Severnogo Ledovitogo okeana v kaynozoye.
Gidrometeorologicheskoye
Izdatel’stvo,
Leningrad,
pp. 34–41.
Kuptsov, V.M., Lisitsin, A.P., 1996. Radiocarbon of quaternary along shore and bottom deposits of the Lena and the
Laptev Sea sediments. Marine Chemistry 53, 301–311.
Lisitzin, A.P., 1996. Oceanic Sedimentation, Lithology and
Geochemistry. American Geophysical Union, Washington,
DC (Chapter 2).
Macdonald, R.W., Bewers, J.M., 1996. Contaminants in the
arctic marine environment: priorities for protection. ICES
Journal of Marine Science 53, 537–563.
Macdonald, R.W., Solomon, S.M., Cranston, R.E., Welch,
H.E., Yunker, M.B., Gobeil, C., 1998. A sediment and
organic carbon budget for the Canadian Beaufort shelf.
Marine Geology 144, 255–273.
Mammone, K.A., 1998. Sediment provenance and transport on
the Siberian Arctic shelf. Masters Thesis, Oregon State
University.
McManus, D.A., Kelley, J.C., Creager, J.S., 1969. Continental
shelf sedimentation in an arctic environment. Geological
Society of America Bulletin 80, 1961–1984.
Meese, D.A., Reimnitz, E., Tucker III, W.B., Gow, A.J.,
Bischof, J., Darby, D., 1997. Evidence for radionuclide
transport by sea ice. The Science of the Total Environment
202, 267–278.
Menke, W., 1984. Geophysical Data Analysis: Discrete Inverse
Theory. Academic Press, London, pp. 1–260.
ARTICLE IN PRESS
1224
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
Munchow, A., Weingartner, T.J., Cooper, L.W., 1998. The
summer hydrography and surface circulation of the East
Siberian Shelf sea. Journal of Geophysical Oceanography
29, 2167–2182.
Naidu, A.S., Creager, J.S., Mowatt, T.C., 1982. Clay mineral
dispersal patterns in the North Bering and Chukchi Seas.
Marine Geology 47, 1–15.
Naugler, F.P., Silverberg, N., Creager, J.S., 1974. Recent
sediments of the East Siberian Sea. In: Herman, Y. (Ed.),
Marine Geology of Oceanography of the Arctic Seas.
Springer, Berlin, pp. 191–210.
Nurnberg,
.
D., Wollenberg, I., Dethleff, D., Eicken, H.,
Kassens, H., Letzig, T., Reimnitz, E., Thiede, J., 1994.
Sediments in Arctic sea ice: implications for entrainment,
transport and release. Marine Geology 119, 185–214.
Pavlov, V.K., 1995. Features of the structure and variability of
the oceanographic processes in the shelf zone of the Laptev
and the East-Siberian Seas. AARI, St. Petersburg, Russia,
Draft.
Pavlov, V.K., Timokhov, L.A., Baskakov, G.A., Kulakov,
M.Yu., Kurazhov, V.K., Pavlov, P.V., Pivovarov, S.V.,
Stanovoy, V.V., 1996. Hydrometeorological regime of the
Kara, Laptev, and East-Siberian Seas. University of
Washington Applied Physics Laboratory Technical Memorandum APL-UW TM 1-96, pp. 79–179.
Pfirman, S., Lange, M.A., Wollenburg, I., Schlosser, P., 1990.
Sea ice characteristics and the role of sediment inclusions in
deep-sea deposition: Arctic–Antarctic comparisons. In:
Bleil, U., Thiede, J. (Eds.), Geological History of the Polar
Oceans: Arctic versus Antarctic. Kluwer Academic Publishers, The Netherlands, pp. 187–211.
Pfirman, S.L., Colony, R., Nurnberg,
.
D., Eicken, H., Rigor, I.,
1997. Reconstructing the origin and trajectory of drifting
Arctic sea ice. Journal of Geophysical Research 102 (C6),
12575–12586.
Pye, K., 1994. Properties of sediment particles. In: Pye, K.
(Ed.), Sediment Transport and Depositional Processes.
Blackwell Scientific Publications, Oxford, pp. 1–24.
Rachold, V., 1999. Major, trace and rare earth element
geochemistry of suspended particulate material of East
Siberian Rivers draining to the Arctic Ocean. In: Kassens,
H., Bauch, H.A., Dmitrenko, I., Eicken, H., Hubberten,
H.-W., Melles, M., Thiede, J., Timokhov, L. (Eds.), Land–
Ocean Systems in the Siberian Arctic: Dynamics and
History. Springer, Berlin, pp. 199–222.
Rachold, V., Hubberten, H.-W., 1999. Carbon isotope
composition of particulate organic material in East
Siberian Rivers. In: Kassens, H., Bauch, H.A., Dmitrenko,
I., Eicken, H., Hubberten, H.-W., Melles, M., Thiede, J.,
Timokhov, L. (Eds.), Land–Ocean Systems in the
Siberian Arctic: Dynamics and History. Springer, Berlin,
pp. 223–238.
Rachold, V., Grigoriev, M.N., Are, F.E., Solomon, S.,
Reimnitz, E., Kassens, H., Antonow, M., 2000. Coastal
erosion vs riverine sediment discharge in the Arctic
shelf seas. International Journal of Earth Sciences 89,
450–460.
Reimnitz, E., Dethleff, D., Nurnberg,
.
D., 1994. Contrasts in
Arctic shelf sea-ice regimes and some implications: Beaufort
Sea versus Laptev Sea. Marine Geology 119, 215–225.
Roach, A.T., Aagaard, K., Pease, C.H., Salo, S.A., Weingartner, T., Pavlov, V., Kulakov, M., 1995. Direct measurements of transport and water properties through the
Bering Strait. Journal of Geophysical Research 100 (C9),
18443–18457.
Rossak, B.T., Kassens, H., Lange, H., Thiede, J., 1999. Clay
mineral distribution in surface sediments of the Laptev Sea:
indicator for sediment provinces, dynamics and sources. In:
Kassens, H., Bauch, H.A., Dmitrenko, I., Eicken, H.,
Hubberten, H.-W., Melles, M., Thiede, J., Timokhov, L.
(Eds.), Land–Ocean Systems in the Siberian Arctic:
Dynamics and History. Springer, Berlin, pp. 587–599.
Schoster, F., Behrends, M., Muller,
.
C., Stein, R., Wahsner, M.,
2000. Modern river discharge and pathways of supplied
material in the Eurasian Arctic Ocean: evidence from
mineral assemblages and major and minor element distribution. International Journal of Earth Sciences 89, 486–495.
Silverberg, N., 1972. Sedimentology of the surface sediments of
the east Siberian and Laptev seas. Ph.D. Thesis, University
of Washington.
Stein, R., Korolev, S., 1994. Shelf-to-basin sediment transport
in the eastern Arctic Ocean. In: Kassens, H., Piepenburg,
D., Thiede, J., Timokhov, L., Hubberten, H.-W., Priamikov, S.M. (Eds.), Berichte zur Polarforschung 144, 87–100.
Stein, R., Fahl, K., 2000. Holocene accumulation of organic
carbon at the Laptev Sea continental margin (Arctic
Ocean): sources, pathways and sinks. Geo-Marine Letters
20 (1), 27–36.
Taylor, J.R., 1982. An introduction to error analysis: the study
of uncertainties in physical measurements. University
Science Books, pp. 1–312.
Timokhov, L.A., 1994. Regional characteristics of the Laptev
and the East Siberian Seas: climate, topography, ice phases,
thermohaline regime, circulation. In: Kassens, H., Piepenburg, D., Thiede, J., Timokhov, L., Hubberten, H.-W.,
Priamikov, S.M. (Eds.), Berichte zur Polarforschung 144,
15–31.
US Naval Oceanographic Office, 1958. Oceanographic Atlas of
the Polar Seas, Part II: Arctic. H.O. Publication No. 705.
Department of the Navy, Washington, DC.
Viscosi-Shirley, C., Mammone, K., Pisias, N., Dymond, J.,
2003. Clay mineralogy and multi-element chemistry of
surface sediments on the Siberian–Arctic shelf: implications
for sediment provenance and grain size sorting. Continental
Shelf Research, doi:10.1016/S0278-4343(03)00091-8.
Wahsner, M., Muller,
.
C., Stein, R., Ivanov, G., Levitan, M.,
Shelekhova, E., Tarasov, G., 1999. Clay–mineral distribution in surface sediments of the Eurasian Arctic Ocean and
continental margin as indictor for source areas and
transport pathways—a synthesis. Boreas 28, 215–233.
Walsh, J.E., 1991. The Arctic as a bellwether. Nature 352,
19–20.
Weingartner, T.J., Cavalieri, D.J., Aagaard, K., Sasaki, Y.,
1998a. Circulation, dense water formation and outflow on
ARTICLE IN PRESS
C. Viscosi-Shirley et al. / Continental Shelf Research 23 (2003) 1201–1225
the northeast Chukchi shelf. Journal of Geophysical
Research 103, 7647–7661.
Weingartner, T.J., Danielson, S., Sasaki, Y., Pavlov, V.,
Kulakov, M., 1998b. The Siberian coastal current: a wind-
1225
and buoyancy-forced Arctic coastal current. Journal of
Geophysical Research 104 (C12), 29697–29713.
Wetherill, B., 1967. Elementary Statistical Methods. Methuen
Inc., London, pp. 1–329.