Basin Research (2007) 19, 87–103, doi: 10.1111/j.1365-2117.2007.00313.x
Large-scale glaciotectonic-imbricated thrust sheets
on three-dimensional seismic data: facts or artefacts?
Bjarne Rafaelsen, n Karin Andreassen, n Kai Hogstadw and Luppo W. Kuilmanz
n
Department of Geology, University of Troms,Troms, Norway
wStatoil Harstad, Harstad, Norway
zNorsk Hydro ASA, Oslo, Norway
ABSTRACT
Imbricate re£ections commonly occur in the glacigenic section of seismic pro¢les from the Bjrnya
Trough.This was the main drainage pathway for fast- £owing ice- streams from the former Barents Sea
and Scandinavian ice sheets. Industry three-dimensional (3D) seismic data from the southern £ank of
the Bjrnya Trough are used here to investigate these imbricate re£ections. Integration of vertical
seismic sections with 3D plan view images and attribute maps reveal that imbricate re£ections at the
SW Barents Sea Margin are mega- scale sediment blocks with a glacigenic origin. Imbricate
re£ections in two regions to the east of the survey appear on plan-view as well-developed lineations of
U- shaped crescents; however, following detailed analysis of their location, geometry and relation to
sailing direction during data acquisition, we can demonstrate that these are seismic artefacts.These
artefacts are related to the straight parts of east^west-trending plough marks on the sea £oor, having a
dip direction that is directly related to the sailing direction of the ship during seismic acquisition. By
analysing both real glacigenic imbrications and false imbrications or artefacts, we are able to
demonstrate the critical distinguishing criterion.
INTRODUCTION
The emergence of three-dimensional (3D) seismic technology as a regional industry exploration tool in the 1990s
has revolutionized seismic investigations. The main advantage of the 3D seismic method over the 2D seismic
method is the dramatic improvement in lateral resolution.
In 3D seismic acquisition, grid spacing is often 25 m or
less, and with advanced 3D seismic migration algorithms
accurate positioning of seismic re£ections is obtained in
all directions, and the Fresnel Zone can be collapsed in
3D.The result can be seen as impressive images of detailed
stratigraphy and complex structures such as thrust-fault
systems and salt domes (Cartwright & Huuse, 2005),
mega- scale glacial lineations (Rafaelsen et al., 2002; Stokes
& Clark, 2002; Andreassen et al., 2004; in press; Evans etal.,
2005; O’Cofaigh et al., 2005) or carbonate build-up complexes (Elvebakk et al., 2002; Rafaelsen et al., 2003). However, it is important not to forget that they are timeimages that may include artefacts.
In this study, 3D seismic technology is used to investigate seismic re£ections that have the appearance of being
glaciotectonic features. These features occur commonly
in the glacigenic section of 3D seismic data from the
southern £ank of the Bjrnya Trough (Fig. 1a). There is
a general consensus that the Bjrnya Trough, the main
cross- shelf trough in the Barents Sea, has been the focus
Correspondence: Bjarne Rafaelsen, Statoil Harstad, PO Box. 40,
9481 Harstad, Norway. E-mail: bjraf@statoil.com
r 2007 The Authors. Journal compilation r 2007 Blackwell Publishing Ltd
of major ice streams draining the Scandinavian and
Barents Sea-Svalbard Ice Sheets during the last glacial
maximum (LGM; Fig. 1a). This is based on bathymetric
considerations and ice- sheet geometry (Denton &
Hughes, 1981), investigations of the Bjrnya Trough
Mouth Fan depocentre of former ice streams (Fig.1a;Vorren & Laberg, 1997), the observation of glacial £utes at
several locations in the Bjrnya Trough (Solheim et al.,
1990; Elverhi et al., 1993), 3D seismic studies of the outer
Bjrnya Trough glacigenic sediments (Andreassen et al.,
2004) and is supported by ice- sheet modelling (Dowdeswell & Siegert, 1999).
2D seismic o¡shore investigations and a shallow borehole have documented that glaciotectonic deformation is
common at the northern £ank of the Bjrnya Trough
(Sttem, 1994) as well as in the Russian SE parts of the
former Barents Sea ice sheet (Gataullin et al., 1993, 2001;
Gataullin & Polyak, 1997), involving unlithi¢ed glacigenic
sediments and often also the upper part of the underlying
consolidated bedrock. Recent ¢eld studies reveal the widespread occurrence and complex nature of glaciotectonic
thrusting on land in northern Russia associated with the
Barents sea and Kara Sea ice sheets (Astakhov, 2001; Larsen et al., 2006).
Semi-regional industry 3D seismic data from the SW
Barents Sea (Fig. 2a; 3D areas 1, 2 and 3) show that largescale imbricate re£ections commonly occur within the
glacigenic sediments that are located at the southern £ank
of the Bjrnya Trough (e.g. Fig. 3b).These data o¡er, with
dense spatial sampling and advanced 3D computer inter-
87
B. Rafaelsen et al.
(a)
Water
depth
(km)
0
0.5
80
76
°N
20°
E
60°E
40°E
°N
Svalbard
Site 986
4.0
Barents Sea
72
Zemly
a
°N
ugh
Novaja
ya Tro
Bjørnø
Bjørnøya
Trough
Mouth
Fan
Scandinavia
K
E
2
1
14
R1
R5
GII
GI
16
sula
p
anin
W
GIII
of
es s
n
i
m
O
wl
flo strea
d
s
e
e
fan
err ic
th
Inf mer
u
o
for
hm
ug
o
r
T
it
s
DP
enin
Russia
(b)
ce
II i
M
LG tent
86
ex
e9
R7
3
URU
20
ESE
WNW
Saale?
24
32
28
NW
SE
(c)
GIII
0.8
Stratigraphic level of Fig. 4
GII
1.0
(s)
R1
Stratigraphic level of Fig. 3
R5
3 km
Fig. 1. (a) Location and bathymetry of the SW Barents Sea. Possible glacier extent and directions of ice stream £ow during Late Glacial
Maximum (28^22 ka BP) and during the retreat of the ice sheet are indicated (modi¢ed from Andreassen et al., 2004). Red box indicates
the position of Fig. 2a. (b) Seismostratigraphic section crossing the study area. Red box indicates three-dimensional (3D) area 1 and
broken red boxes indicate the extrapolated position of 3D seismic areas 2 and 3. Not to scale. Approximate vertical scale is 1200 m.
Modi¢ed fromVorren et al. (1990). (c) Seismic pro¢le from 3D area 1.
pretation techniques, a fundamentally new method for
studying such features. In this paper, we investigate these
imbricate re£ections and their 3D appearance. In 3D area
1, we discuss the imbricate re£ections glaciological significance and relationship with other associated geomorpho logical features, and put constraints on their origin and
formation. In 3D area 2 and 3, imbricate re£ections may
88
easily be mis-interpreted as glaciotectonic thrust sheets
and thereby represent a pitfall for the interpreter. These
imbricate re£ections represent a 3D seismic artefact that
has not been described before, and here we describe how
this artefact appears in three dimensions. We also reveal
how it is related to the geomorphology on the sea £oor
and the 3D seismic acquisition geometry.
r 2007 The Authors. Journal compilation r 2007 Blackwell Publishing Ltd, Basin Research, 19, 87^103
Glaciotectonic thrust sheets: facts or artefacts?
0
(a)
N
74°N
0.5
1
km below
sea level
gh
rou
ya T
rnø
Bjø
2
3
72°N
Nordkappbanken
1
30°E
Ingøydjupet
nd
ainla
ian m
eg
Norw
20°E
and
s
land
is
100 km
25°E
N
(b)
3D-A
72.5° N
3
−0.2
−0.3
2
−0.4
−0.5
72° N
km below
sea level
10 km
21° E
23° E
Fig. 2. (a) Map of the sea £oor ( 20 vertical exaggeration) revealing how the palaeo -ice £ow could ¢t with the direction of the imbricate
re£ections dipping eastwards in three-dimensional (3D) area 2 and 3. Note the location of the 3D areas relative to large- scale elements
on the sea £oor. Black lines indicate palaeo -ice £ow. Arrows indicate the main direction of the imbricate re£ections. Depth is in
kilometres below sea level and the grey circle shows the position of the light source.The map is a Statoil compilation of bathymetry
based on seismic data, compiled by NGU/Eirik Mauring. A publication on details from the sea £oor in the Barents Sea is presented in
Andreassen etal. (2007). For location of map, see Fig.1a. (b) Detailed map of the 3D- surveys 2 and 3 with the location, orientation, length
and dip direction of the observed imbricate re£ections. Note how the imbricate re£ections in 3D area 2 occur in 4^10 km wide eastwards
or westwards dipping belts.
r 2007 The Authors. Journal compilation r 2007 Blackwell Publishing Ltd, Basin Research, 19, 87^103
89
B. Rafaelsen et al.
N
(a)
0
RMS-Amplitude
21222
2 km
(b)
1.2
0.5 km
1.4
(s)
N
(c)
0
RMS-Amplitude
19165
2 km
Fig. 3. (a) Root mean square (RMS) amplitude map from the lower part of unit GII showing the orientation of mega- scale glacial
lineations (arrow) in the upper shaded (green) volume in (b) [three-dimensional (3D) area1; Fig1c].The small black rectangle to the lower
left represents the 3D seismic survey and the white rectangle within represents the area shown in this ¢gure. (b) Seismic pro¢le with
thrust sheets on two stratigraphic levels within GII, where the green and blue colours represent the zones used when calculating the
RMS amplitude maps. (c) RMS amplitude map with the position and size of the sediment blocks in the lower shaded (blue) volume in
(b).The small black rectangle to the lower left represent the 3D seismic survey and the white rectangle within represents the area shown
in this ¢gure. Modi¢ed from Andreassen et al. (2007).
PHYSIOGRAPHIC SETTING AND
SEISMIC STRATIGRAPHY
The Barents Sea is an epicontinental sea characterized by
relatively shallow banks separated by deep troughs, which
have been excavated by glacial erosion.The bank areas have
water depths of o300 m, whereas the most pronounced
trough, the Bjrnya Trough, reaches depths of 500 m.
90
The Bjrnya Trough ice stream developed in an area of
strongly convergent ice £ow from the Svalbard-Barents
Sea and the Scandinavian ice sheets (Fig.1a). It is modelled
to have £owed at about 800 m a 1 and produced about
30 km3 a 1 of icebergs during full glacial periods (Siegert
& Dowdeswell, 2002). The Bjrnya Trough Mouth Fan,
located at the mouth of this trough, acted during the
Plio -Pleistocene as a main depocentre for ice streams that
r 2007 The Authors. Journal compilation r 2007 Blackwell Publishing Ltd, Basin Research, 19, 87^103
Glaciotectonic thrust sheets: facts or artefacts?
drained to the Bjrnya Trough shelf break at glacial maxima (Fiedler & Faleide, 1996; Kuvaas & Kristo¡ersen, 1996;
Vorren & Laberg, 1996; Andreassen et al., 2004).
Mega- scale glacial lineations are parallel- elongated
groove-ridge structures and they occur in the study area
on buried surfaces within the glacigenic succession
(Fig. 4a; red, blue and purple arrows). The mega- scale
glacial lineations in the study area have a relief of up to
10 m, widths from 50 to 360 m, length of 38 km and an elongation ratio of up to 105: 1 (Andreassen et al., 2004). Such
features are interpreted to have been formed subglacially
in areas of fast ice £ow (Stokes & Clark, 2002), and indicate
the direction of palaeo -ice £ow.
Three main sediment packages, GI, GII and GIII (Fig.
1b), have been identi¢ed and correlated along the western
Barents Sea^Svalbard margin (Faleide et al., 1996; Butt
et al., 2000). ODP Site 986 west of Svalbard (Fig. 1a) is a
key borehole for age constraints of the seismic stratigraphy
of this area (Forsberg et al., 1999; Butt et al., 2000); additional chronological constraints come from exploration
wells (Eidvin et al., 1993,1998) and shallow borings (Sttem
et al., 1992; Sttem, 1994). It is, however, a problem that
biostratigraphic data used for age estimates are often retrieved from drill cuttings, where sediment collapse in
the borehole adds younger material to the samples. In addition, glacigenic sediments may contain high amounts of
older reworked material.
The regional seismic re£ection, R7, which can be followed from ODP Site 986 southward along the margin to
the base of sediment package GI of the Bear Island
Trough Mouth Fan (Fig. 1b), has an age estimate of
2.3 Ma (Forsberg et al., 1999). Re£ection R5, at the base of
package GII (Fig.1c), represents the ¢rst documented occurrence of grounded glaciers reaching the Bjrnya
Trough shelf break (Andreassen et al., 2004), and is given
the interpolated age estimate of 1.6^1.7 Ma from ODP
Site 986 - data (Butt et al., 2000), whereas Faleide et al.
(1996) suggested a likely age of about 1.0 Ma. Amino acid
analysis of R1, at the base of GIII, indicates an age younger than 440 ka (Sttem et al., 1992), whereas the extrapolation of calculated sedimentation rates in piston cores
on the Svalbard margin has given an approximate age
of 200 ka (Elverhi et al., 1995). A likely age of R1 is thus
between 440 and 200 ka.
An upper regional unconformity (URU; Fig. 1b) developed over the entire Barents Sea shelf as a result of repeated glacial erosion (e.g. Solheim & Kristo¡ersen, 1984;
Vorren et al., 1986; Lebesbye, 2000; Rafaelsen et al., 2002). It
most likely represents the erosional base for several continental shelf glaciations; however, the seismic stratigraphy
suggests that the last major erosion down to the level of the
URU occurred at the SW Barents Sea shelf at R1 time, and
that the sediments overlying the URU in this area are
younger than 440 ka (Faleide et al., 1996; Kuvaas & Kristoffersen, 1996). The URU on the shelf represents a change
from an early erosional glacial regime to a later aggradational regime, whereas the stratigraphy above indicates
that ice sheets have reached the SW Barents Sea shelf edge
at least ¢ve times (Vorren et al., 1991) since the URU was
formed.
3D SEISMIC DATA
This study is based on three industry 3D seismic data sets
located at the southern £ank of the Bjrnya Trough
(Fig. 2a; 3D areas 1, 2 and 3), covering a total of 4000 km2.
The data sets are located at water depths of 150^400 m.
The surveys were acquired in 1998 with a line spacing of
up to 37.5 m (Table 1). The dominant frequency of the data
is around 40 Hz at the studied depths.The theoretical vertical resolution of the data, using 1750 m s 1 as an average
velocity of the glacigenic sediments, is therefore around
10 m (1/4 of the dominant wavelength). In practice, the
occurrence of thin beds and associated interference e¡ects
can subtly distort the relief of two -way time picks. In addition, 3D seismic horizons represent an average response
over at least 1/4 of a wavelength, possibly up to half a wavelength (Bulat, 2005), giving a vertical resolution of 10 m
(1/4 of the dominant wavelength) to 20 m (1/2 of the dominant wavelength).
The theoretical lateral resolution is de¢ned by the
width of the Fresnel zone, which, after 3D migration, is
also around 1/4 of the dominant wavelength (Brown,
2005), i.e.10 m. Errors in migration may reduce the practical resolution to 1/2 the dominant wavelength (Brown,
2005; Bulat, 2005).The spatial sampling of the seismic volume is the primary control of horizontal resolution (Cartwright & Huuse, 2005), as the spatial sampling is generally
coarser than or equal to the dominant wavelength of the
signal (Bulat, 2005). In this study, the bin spacing and
therefore the horizontal resolution is up to 37.5 m. 2D seismic lines and wells have been used for stratigraphic correlation with other work in the area.
DIPPING REFLECTIONS IN 3D SEISMIC
AREA 1
The glacigenic sediment packages GII and GIII contain
more than 900 m of palaeo - shelf sediments due to subsidence in 3D area 1. Long chains of megablocks and rafts,
buried in thick till units between glacially eroded horizons, occur commonly within the shelf units of the sediment packages GII and GIII in 3D area 1 at the Barents
Sea margin (Fig. 1c). These sediment blocks have previously been documented by Andreassen et al. (2004,
2007), and are shown here in Figs 3 and 4. Sediment blocks
larger than 1km2 are classi¢ed as medium- sized megablocks whereas those smaller than 1km2 are termed rafts
(based on Aber et al., 1989). Elongated groove-ridge structures, termed mega- scale glacial lineations, have a relief of
up to 10 m, widths from 50 to 360 m and lengths exceeding
38 km (Andreassen et al., 2004). High-amplitude seismic
anomalies, interpreted as mega-blocks with di¡erent
acoustic properties than the surrounding sediments, are
r 2007 The Authors. Journal compilation r 2007 Blackwell Publishing Ltd, Basin Research, 19, 87^103
91
B. Rafaelsen et al.
N
(a)
639
ms
B
D
989
ms
0.7
(b)
0.8
(s)
2 km
1.0 km
N
(c)
D
B
(d)
0.7
0.8
2 km
RMS-Amplitude
0
17796
(s)
0.5
kmkm
1.0
N
(e)
B
2 km
0
D
RMS-Amplitude
17796
Fig. 4. Small black rectangle in the upper left corner represents the three-dimensional (3D) seismic survey and the white rectangle
within represents the area shown in that ¢gure. (a) Shaded relief map of an intra GIII horizon (see Fig. 1c) underlying the sediment
blocks in (b). Note the mega- scale glacial lineations (yellow arrows) oriented southwest/northeast (blue arrow).The position of the light
source is shown by the grey circle (b) Seismic pro¢le with mega- scale glacial lineations. (c) Volumetric Root Mean Square map (RMS;
shaded volume) with chains of sediment blocks from intra GIII (see Fig.1c). Note that the orientation of the sediment chains is similar to
that of the mega- scale glacial lineations in a. Modi¢ed from Andreassen et al. (2004). (d) Seismic pro¢le reveals related thrusts. (e) RMS
amplitude map (b) draped on top of the shaded relief map (a) showing two generations of mega- scale glacial lineations (1 and 2)
combined with the sediment blocks. Modi¢ed from Andreassen et al. (2007).
92
r 2007 The Authors. Journal compilation r 2007 Blackwell Publishing Ltd, Basin Research, 19, 87^103
Glaciotectonic thrust sheets: facts or artefacts?
Table1. Parameters used in the 3D seismic acquisition
Weather during acquisition
Acquired
Source depth
Receiver depth
Source interval
Line spacing
Group interval
Sampling interval
Low- cut ¢lter
High- cut ¢lter
3D area 1
3D area 2
3D area 3
Very good
1998
6m
8 1m
25 m
37.5 m
12.5 m
2 ms
3 Hz^12 dB/oct
206 Hz^76 dB/oct
Very good
1998
6m
5 1m
25 m
25 m
12.5 m
2 ms
3 Hz^18 dB/oct
180 Hz^72 dB/oct
Very good
1998
6m
7 1m
25 m
37.5
12.5 m
2 ms
3 Hz^18 dB/oct
180 Hz^72 dB/oct
Time-variant ¢lters have later been applied to the 3D surveys, signi¢cantly reducing the high- cut frequency values displayed here during the subsequent
processing of the data.
3D, three dimensional.
observed at the base of GII (Fig. 3). Root mean square
(RMS) volumetric attribute maps show a lateral distribution of sediment blocks and the appearance of mega- scale
glacial lineations (Fig. 3a), whereas seismic pro¢les show
their stacked glaciotectonic nature (Fig. 3b). The megascale glacial lineations indicate the direction of palaeo -ice
£ow (Fig. 3a).
On the interpreted seismic horizon Intra GIII, and in
seismic pro¢les, mega- scale glacial lineations are evident
(Fig. 4a and b). They are oriented in the same direction as
the chains of sediment blocks (Fig. 4c and d).The chains of
mega-blocks and rafts are aligned in over 50-km long
chains that are 1^2 km wide (Andreassen et al., 2004) and
have the same NE-trending orientation as the observed
mega- scale glacial lineations.
Seismic pro¢les parallel to the orientation of the sediment chains and the inferred ice £ow indicate, where the
vertical resolution is good enough, that the mega- scale
blocks consist of back-tilted sub-horizontal sediment
slabs that have been displaced from the NE along a series
of shear planes (Figs 3b and 4d), parallel to the inferred icestream £ow. An image, obtained by calculating the RMS
attribute for a volume that includes the horizon from Fig.
4a and the sediment blocks (Fig. 4b; shaded zone), has
been draped over the interpreted shaded relief map (Fig.
4a) and reveals the relationship between the glacial lineations and mega-blocks (Fig. 4e). This combined attribute/
shaded-relief image of Fig. 4e shows that the sediment
blocks have been cut by two sets of mega- scale glacial
lineations, suggesting that ice streams have eroded the already detached mass of sediment blocks after they were
deposited.
Thrust faults related to the mega-blocks and rafts disturb the upper or entire section of the sediments between
more pronounced seismic horizons above URU. These
stacked mega-blocks dip towards the NE (Figs 3 and 4), indicating a glacier movement towards the south-west, as
documented by Andreassen et al. (2004).
The chains of sediment blocks are parallel to megascale glacial lineations and the inferred £ow of ice streams.
The common occurrence of mega- scale sediment blocks
and rafts within sediment packages GII and GIII suggests
that glaciotectonic erosion, transport and deposition by
ice streams accounts for the high sediment £ux from the
Barents Sea to the shelf break and the Bear Island Trough
Mouth Fan since the early Pleistocene.
IMBRICATE REFLECTIONS IN 3D
SEISMIC AREAS 2 AND 3
The glacigenic sediments in 3D seismic areas 2 and 3 are
separated from underlying dipping, consolidated sediments by the URU (Fig. 1b). In this area, the thickness of
the glacigenic sediments is up to 150 m.The glacigenic sequence is commonly disturbed by imbricate re£ections
that occur over large parts of these two 3D areas.They occur on vertical seismic pro¢les as dipping re£ections that
are stacked obliquely in an imbricated manner (Figs 5a and
6a, c).The dipping re£ection segments are each 200^300 m
wide (Figs 5b, c and 6b, d), 300^850 m long, 22^100 m in
vertical extent and inclined between 2.51 and 221 (mainly
81^121). They occur in patches that reach a total length of
0.5^5 km.The imbricate re£ections reach down to URU or
more frequently the horizon just above URU (Figs 5a
and 6a, c).
The integration of vertical seismic sections with mapview images shows that the imbricate re£ections run parallel to the straight parts of the £anks of east^west-trending plough marks on the sea £oor (Fig. 7). The dip
directions of the imbricate re£ections are mainly towards
the east (Figs 2b and 6a; totally 70 structures) or west (Figs
2b and 6c; totally 76 structures). In some instances less distinct, almost semi-transparent/lower-amplitude areas
have been observed on the seismic pro¢les on the stossside of the imbricate re£ections (Fig. 8a). The observed
structures have been mapped on all glacigenic horizons
within 3D areas 2 and 3.
The imbricate re£ections observed in 3D surveys 2 and
3 appear on time slices (Figs 5b, c and 6b, d) and horizon
r 2007 The Authors. Journal compilation r 2007 Blackwell Publishing Ltd, Basin Research, 19, 87^103
93
B. Rafaelsen et al.
(a)
0.6
(s)
-
(b)
50 m
0.5
URU
1 km
(c)
496 ms
N
524 ms
N
Fig. 5. (a) Seismic pro¢le from three-dimensional (3D) area 2 with imbricate re£ections that sole out along a decollement zone above
the Upper Regional Unconformity (URU). (b^ c) The time slices show the characteristic smooth U- shape and the linearity of the
imbricated thrust lineations. A slight increase in size of the U- shape with depth can be measured using the interpretation software, but
is di⁄cult to observe with the naked eye.The small black rectangle in the lower left corner represents the 3D seismic survey and the
white rectangle within represents the area shown in that ¢gure.
(a)
(c)
0.45
50 m
0.5
URU
0.6
(s)
0.55
0.5 km
(b) N
520 ms
(d)
0.5 km
0.5 km
(s)
N
504 ms
0.5 km
Fig. 6. (a and c) Seismic pro¢les from three-dimensional (3D) area 2 with east dipping (a) and west dipping (c) imbricate re£ections in
the upper part of the glacigenic succession.The thickening of the sequence in (c) may in 2D appear to be related to the imbricate
re£ections, but they are genetically unrelated to each other. (b and d) Time slices illustrating their imbricated horizontal appearance.
The small black rectangle in the upper right corner represents the 3D seismic survey and the white rectangle within represents the area
shown in that ¢gure.
94
r 2007 The Authors. Journal compilation r 2007 Blackwell Publishing Ltd, Basin Research, 19, 87^103
Glaciotectonic thrust sheets: facts or artefacts?
146 mapped imbricate re£ection patches indicated in 3D
areas 2 and 3 on Fig. 2b are those that are large enough to
be visible on time- structure maps, but in addition many
vertically less extensive U- shaped structures, only detectable on time slices and amplitude maps, have also been
observed.
(a)
N
1 km
0.5 (b)
0.6
(s)
0.5 km
(c)
N
1 km
(d)
N
0.5 km
Fig. 7. (a, c and d) Illuminated shaded relief maps of the sea £oor
in three-dimensional (3D) area 3, draped with a horizon slice
amplitude map showing amplitude variations 4 ms below the sea
£oor. U- shaped imbricate re£ections occur parallel to straight
parts of the east^west-trending plough marks.The position of
the light source is shown by the grey circle. For location of a, see
Fig. 11b.The small black rectangle in the upper right corner
represents the 3D seismic survey and the white rectangle within
represent the area shown in that ¢gure. (b) Seismic pro¢le with
imbricate re£ections. Dark colours represent strong positive
amplitudes.
maps (Fig. 9) as smooth U- shaped features stacked linearly one after another and with equal widths.The imbricate
re£ections occur on time slices and shaded relief maps of
buried horizons throughout 3D area 3, sometimes crosscutting mega- scale glacial lineations (Fig.9). In Fig.9 their
orientation is trending east^west, whereas the mega- scale
lineations generally trend north^ south. In 3D area 2, the
dipping re£ections are clustered in 4^10-km-wide belts
of structures dipping either westwards or eastwards (Fig.
2b). Their concentration is highest in the middle section
of the 3D area, declining northwards and southwards. The
INTERPRETATION AND DISCUSSION
Glacial megablocks and rafts are common in formerly glaciated areas, and may be aligned in chains (Aber et al.,1989).
The occurrence of cupola hills buried within a till unit
have been documented at the northern £ank of Bjrnya
Trough by Sttem (1994) in a unit that correlated with the
very upper part of sequence GIII. One of the cupola hills
consisted of a 15^25-m-thick block of Cretaceous sedimentary rock, buried in the Pleistocene till. A cupola hill
has a glaciotectonic origin and has an overall dome-like
morphology shape (Aber et al., 1989).
3D seismic images of the buried surfaces are within sediment packages GII and GIII in 3D area 1 characterized
by mega- scale glacial lineations that extend for over
38 km over the entire survey area. The dominant orientation of these lineations is generally north- easterly (Figs 3a,
c and 4a, e), which is consistant with orientation of ice
streams draining out the Bjrnya Trough (Fig. 1a). Imprints from ice streams that drained to the shelf edge during the LGM are clearly visible on the large- scale shaded
relief image of the SW Barents Sea sea£oor (Fig. 2a). The
mega- scale sediment blocks and the other thrust-features
observed in 3D area1 (Figs 3 and 4) are, from their orientation between glacially eroded surfaces and from their
internal structural consistency with mega- scale glacial
lineations, interpreted to have been subjected to deformation by glaciers draining out the Bjrnya Trough. Andreassen et al. (2004) suggested this was most probably
due to fast- £owing ice streams. The detached sediment
blocks are interpreted to have been picked up, transported
and deposited by glaciers. Transportation of glacial megablocks is normally interpreted to be formed by basal freezing onto the base of the glacier (Aber et al., 1989). Andreassen et al. (2004) propose that palaeo -ice streams draining
through the Bjrnya Trough could have undergone periods of basal freezing near their margins, caused by a fast
downward advection of cold surface ice. The frozen sediment blocks are interpreted to have been entrained by the
glacier and transported along thrust planes before being
re-deposited at the margin. Basal freezing is known to
cause over- consolidation of sub-glacial sediments and
can produce a rheological contrast between the frozen sediments at the ice base and the less- consolidated sediments with depth. This may focus the thrust deformation
at the transition between the over- consolidated and lessconsolidated sediments.
Large- scale glaciotectonic thrust sheets similar to
those described from 3D area 1 have been described from
former glaciated areas on land in NW Europe and North
r 2007 The Authors. Journal compilation r 2007 Blackwell Publishing Ltd, Basin Research, 19, 87^103
95
B. Rafaelsen et al.
(a)
50 m
0.5
0.6
(s)
− 32768
0.5 km
32767
URU
(b)
N
U-shape
0.5 km
< 100 m
(c)
500
0m
30
m
Apparent
basal
decollement
zone
Inferred Ice
advance
Fig. 8. (a) Vertical seismic pro¢le showing several kilometres of well-developed imbricate re£ections dipping towards the east [threedimensional (3D) area 3].The blue line indicates the location of the time slice in (b).The seismic pro¢le has been ¢ltered, post-stack,
with a low- cut of 40 Hz. (b) Time slice at 572 msTWTwith U- shaped imbricated thrust lineations.The small black rectangle in the
upper right corner represents the 3D seismic survey and the white rectangle within represents the area shown in that ¢gure. (c)
Constructed model of 3D appearance of the glaciotectonic imbricated features.
America (Clayton et al., 1980; Aber et al., 1989; Huuse &
Lykke-Andersen, 2000). In Denmark, cli¡- exposures on
land have allowed detailed sedimentological and structural
studies of such thrust sheets (Pedersen, 1987, 2000; Klint
& Pedersen, 1995). At Mns Klint the thrust sheets are ca.
100 m thick and 1km long, and at Lnstrup Klint the
stacked imbricated thrust sheets are ca. 4.5 km long. Imbricated glaciotectonic structures from push-moraines have
been described from Svalbard (Van der Wateren, 1995;
Boulton etal.,1999). Elsterian/Saalian glaciotectonic thrust
structures from marine areas, studied in two dimensions
96
using 2D seismic data, are 200 m long, 100^250 m thick
and interpreted to have been formed by gravity spreading
in front of advancing ice sheets (Huuse & Lykke-Andersen,
2000; Andersen et al., 2005). 2D seismic and shallow
boreholes have been used for the identi¢cation of a glacio tectonized bedrock surface (Sttem et al., 1992) and glacio tectonic structures such as hill-hole pairs (Sttem, 1990).
Also, glaciotectonic features imply strong glacial erosion
of glacigenic sediments (Gataullin et al., 2001; Larsen et
al., 2006) and Mesozoic rocks in the eastern Barents Sea
(Gataullin et al., 1993).
r 2007 The Authors. Journal compilation r 2007 Blackwell Publishing Ltd, Basin Research, 19, 87^103
Glaciotectonic thrust sheets: facts or artefacts?
lac
ial
line
a
tion
s
N
Me
ga
-sc
a
le g
Imbricate
reflections
486.31
637.77
1 km
ms below sea floor
Fig. 9. Illuminated shaded relief map of a buried horizon above the upper regional unconformity in three-dimensional (3D) area 3
showing imbricate re£ections and mega- scale glacial lineations. Note how the imbricate re£ections here crosscut the direction of the
mega- scale glacial lineations.The upper encircled structure is shown in the seismic pro¢le on Fig. 11a.The grey circle shows the
position of the light source.The small black rectangle in the upper right corner represents the 3D seismic survey and the white rectangle
within represent the area shown in the ¢gure.
(a)
Formation theory
Hanging wall
Foot wall
anticline
syncline
(b)
Ice
Ice
Decollement level
Youngest
Oldest
Vertikal limit
3
2
1
Youngest
Oldest
Vertikal limit
6
5
4
Old, overridden
thrust blocks
3
2
1
Fig. 10. (a) Schematic diagram illustrating continued displacement of inclined ice-marginal thrust blocks (4, 5, 6). (b) The older thrust
blocks are probably eroded in the upper parts when the glacier overrides them (1,2, 3). Not to scale. Modi¢ed from Aber et al. (1989).
The imbricate re£ections of 3D areas 2 and 3 (Figs 5^7)
show many of the same characteristics as those of 3D area
1.They appear in map view as cresentic U- shaped features
(Figs 5b, c, 6b, d and 7a, c, d), a characteristic that has been
described for both large- and small- scale glaciotectonic
ridges (Aber et al., 1989).The geological history of the study
area with repeated glacials/inter-glacials since the formation of URU has directed us towards a glaciotectonic origin for the observed imbricated features.
Decollement zones, in connection with glaciotectonics,
are primarily dependent on lithology (Pedersen, 1996).
Marine clay can act as a decollement zone for the thrust
sheets and, because of increased pore pressure, be plastically deformed (Pedersen, 1987; Huuse & Lykke-Andersen,
2000). A duplex of imbricate thrust sheets develop as the
sediments are deformed and overthrusted, which are
among characteristic structures produced by gravity
spreading (Pedersen, 1987). This may occur when faults
form thrust blocks underneath and in front of a glacier
and place them in an imbricated relation to each other
(Pedersen, 2000). In this way, imbricated structures
are constructed (Fig. 10) in which proximal thrust blocks
r 2007 The Authors. Journal compilation r 2007 Blackwell Publishing Ltd, Basin Research, 19, 87^103
97
B. Rafaelsen et al.
(a)
(b)
N
Seismic amplitude
− 28522
32767
2 km
Fig. 7a
(c)
0.50
1 km
0.55
592 ms
Fig. 11. (a^b) Amplitude map 4 ms below the sea£oor in three-dimensional (3D) area 3, with white arrows indicating the sailing
direction during acquisition of the 3D seismic data. Note how the U- shape of the imbricate re£ections dominates related to the sailing
direction.The small black rectangle represents the 3D seismic survey and the white rectangle within represents the area shown in the
¢gure. (c) Vertical seismic pro¢le of imbricate re£ections dipping eastwards and a horizontal time slice revealing part of the
characteristic U- shape. Note the linearity and how smooth and systematic the U- shape appears in the time slice. Location in (b).
(oldest) may be moved several kilometres, whereas the
most distal (youngest) will move only a short distance
(Aber et al., 1989). Thrust structures may also form englacially (Hambrey et al., 1997), but they are generally steeper
(20^351) and less extensive (10^100 m wide and long) than
those described from 3D area 2 and 3.
After a close examination of the 3D seismic data, an image of the 3D appearance of the imbricate re£ections in
areas 2 and 3 was constructed (Fig. 8c). The following observations may suggest that the imbricate re£ections could
be glaciotectonically formed: (i) the imbricate re£ections
occur with a varying angle to the inline direction (Fig.
2b), (ii) the re£ections are only observed in the glacigenic
succession, (iii) the imbricate re£ections do not seem to
be associated with the orientation of the acquisition
98
(Fig. 2b), (iv) the imbricate re£ections tend to die out along
a decollement plane (Fig. 8a), (v) in some cases a thickening
of the sequence related to the imbricate re£ections is observed, as expected, due to compression (Fig. 6c). If the imbricate re£ections represent compressed and more
consolidated sediment than sediments surrounding them,
it is possible that icebergs drifting with their keel down
into the sediments, creating turning and bending plough
marks, were sometimes pushed against imbricated features and forced to move along them until the prevailing
wind and current direction changed or until it passed the
imbricated features.
On the other hand, the following has puzzled the
authors: (i) The geographical distribution (Fig. 2b) of imbricate re£ections dipping towards the east (Fig. 6a) and
r 2007 The Authors. Journal compilation r 2007 Blackwell Publishing Ltd, Basin Research, 19, 87^103
Glaciotectonic thrust sheets: facts or artefacts?
N
Water
depth
(in ms)
455
610
10 km
=
=
Fig. 12. Time- structure map of the sea £oor in threedimensional (3D) area 2. Yellow arrows indicate the sailing
direction of the vessel during acquisition. Black and white arrows
show the orientation, dip direction and length of the imbricate
re£ections (black arrows indicate eastwards dip, white arrows
indicate westwards dip).There is almost a1 : 1correlation between
sailing direction during acquisition and the dip direction of the
imbricate re£ections.
west (Fig. 6c).The occurrence of imbricate re£ections with
an easterly dip ¢ts well within a model of glaciotectonic
deformation in the convergence zone of fast- £owing ice
streams draining out Ingydjupet and Bjrnya Trough,
and is in consistence with imprints of mega- scale glacial
lineations on the sea £oor (Fig. 2a). The imbricate re£ections with a westerly dip do not, however, ¢t into such a
glacigenic setting and have been di⁄cult to explain from
a geological point of view. Also, the alternating zones of
westerly and easterly dipping thrust sheets in 3D area 2
have been di⁄cult to explain (Fig. 2b). (ii) There is no correlation between the orientation of imbricate re£ections
and mega- scale glacial lineations on buried surfaces
(Fig. 9; Rafaelsen et al., 2002), the imbricate re£ections
crosscut the mega- scale glacial lineations. (iii) On mapview images, the smooth symmetrical appearance of the
U- shape of the imbricate re£ections (Fig. 5b and c), their
amazingly even width of around 200^300 m persisting for
up to 5 km length (Fig. 8b), and the extreme linearity of
the stacked structure (Fig. 9) is much more organized and
less rough and chaotic than expected. (iv) Signi¢cant disturbance of the sediments laterally and proximally of the
structures is generally lacking. (v) The imbricate re£ections occur along straight parts of the edge of east^westtrending plough marks, but disappear where the plough
mark bends (Fig. 11), i.e. where the direction of iceberg
transports change. In addition, the imbricate re£ections
never occur along the straight north^ south-trending parts
of plough marks (Fig. 11). The relationship between the
straight parts of iceberg plough marks (Fig. 7) and imbricate re£ections was not obvious until sea £oor maps were
combined with amplitude maps from below the sea £oor
(Fig. 11). (vi) On seismic pro¢les the imbricate re£ections
possess similar regular dips, regular spatial frequency
and crosscut re£ectors. (vii) In plan view the features appear with a regular spatial period and are very linear.
A breakthrough came when we discovered that there is
a one-to - one correlation between the sailing direction
during seismic acquisition and the dips of the imbricate
re£ections within both 3D area 2 (Fig. 12) and 3D
area 3 (Fig. 13). This correlation indicates that the thrust
features in 3D areas 2 and 3 are artefacts. One way to get
an idea of the sailing direction during 3D seismic acquisition is to study the edges of the data set. To get full-fold
within the area of interest, the ship must start acquisition
before entering the area of interest. Therefore, more data
normally occur on the eastern side of the survey when the
ship sails westwards, and when the ship sails eastwards
more data occur on the western side of the survey (Figs 12
and 13). However, 3D seismic surveys are in some cases cut
at the edges showing only the area of interest and not the
data without full fold, and in these cases this method
cannot be used.
Further investigations revealed that the red^white^
black compressed colour scale used when viewing seismic
pro¢les may cause the imbricated features to appear as soling out along a decollement zone, whereas grey-scale uncompressed images in some cases reveal that the features
propagate further down beyond URU and into the Cretaceous and Triassic part of the succession before they weaken
beyond detection (Fig. 14). Additionally, the smooth
U- shape of the imbricated features widens slightly in plan
view with depth (Fig. 5).These observations illustrate that
the use of correct colour scale is important in the interpretation of seismic data, particularly 3D seismic data.
In relation to the sailing direction of the ship during acquisition, the imbricated features tend to occur on the ‘leeside’of plough marks (i.e. the opposite side from which the
ship came; Fig. 11). In some cases, weak imbrication features also occur on the plough marks ‘stoss- side’ relative
to the acquisition direction (Fig. 7c). The cause of this
preferential occurrence is not clear.
r 2007 The Authors. Journal compilation r 2007 Blackwell Publishing Ltd, Basin Research, 19, 87^103
99
B. Rafaelsen et al.
N
Water
depth
(in ms)
397
=
10 km
=
574
Fig. 13. Time- structure map of the sea £oor in three-dimensional (3D) area 3. Yellow arrows indicate the sailing direction of the vessel
during acquisition. Black and white arrows show the orientation, dip direction and length of the imbricate re£ections (black arrows
indicate eastwards dip, white arrows indicate westwards dip).There is almost a 1 : 1 correlation between sailing direction during
acquisition and the dip direction of the imbricate re£ections.
(a)
(c)
0.5
0.6
URU
URU
0.7
− 32768
0.5 km
32767
0.5 km
− 32768
32767
(s)
(d)
(b)
0.5
0.6
URU
URU
0.7
0.5 km
− 32768
(s)
32767
0.5 km
− 32768
32767
Fig. 14. Vertical seismic pro¢les showing that the appearance of imbricate re£ections depends on the display.The small black rectangle
in the upper right corner represents the three-dimensional (3D) seismic survey and the white rectangle within represents the line
shown in that ¢gure. (a and c) Seismic pro¢les from 3D area 3 viewed with a compressed red^white^black colour scale. Note how the
imbricate re£ections appear to weaken and possibly sole out before reaching upper regional unconformity (URU). (b and d) The same
seismic pro¢les, as in (a and c), are here viewed with an uncompressed grey- scale.The imbricate re£ections extend deeper, even beyond
URU, and do not to sole out.
The imbricate re£ections in 3D areas 2 and 3 are based
on the evidence above interpreted to be related to plough
marks on the sea £oor (Fig. 7). The imbricate re£ections
may represent aliased energies associated with the plough
100
marks or di¡raction hyperbolas set up by the edges of the
linear parts of iceberg plough-marks parts. The fact that
these imbricate re£ections appear as U- shapes and not
circles in time slices are unclear, but could be related to
r 2007 The Authors. Journal compilation r 2007 Blackwell Publishing Ltd, Basin Research, 19, 87^103
Glaciotectonic thrust sheets: facts or artefacts?
the acquisition geometry or seismic processing. These artefacts are either di⁄cult to remove during processing or
not thoroughly dealt with as they are located in much shallower depths than the target depth of the 3D seismic surveys.These results reveal the importance of consulting the
acquisition parameters when interpreting 3D seismic data.
CONCLUSIONS
Integration of vertical seismic transects with 3D shaded
relief images of buried surfaces, volumetric attribute maps
and time slices provides an excellent tool for detecting and
imaging mega- scale thrust sheets, and for understanding
the processes involved in glaciotectonic deformation.
Glaciotectonic features as mega- scale blocks and rafts
are common within the upper two sediment packages GII
and GIII in 3D area 1. These glaciotectonic features are
parallel to mega- scale glacial lineations, which suggest
erosion by fast- £owing ice streams during Pleistocene glaciations.
The level of detail in the 3D seismic images can, however, seduce the interpreter into treating the image purely
as an aerial photograph (Bulat, 2005), and it is important
not to forget that they are time-images that may include
artefacts. The excellent correlation between shooting direction and dip of apparent thrust- sheet features in 3D
areas 2 and 3 indicates that the apparent thrust features in
these two areas are artefacts and not real. These dipping
re£ections are interpreted to be related to the £anks of iceberg plough marks on the sea £oor. In plan view images
such as time slices, and shaded relief images of interpreted
horizons, they appear as lineations of smooth U- shaped
crescents.
3D seismic data allow us to visualize large areas of the
subsurface in unprecedented detail, and may reveal geolo gical features that have not been described before. The
data might, however, also contain 3D artefacts that have
not been described before and which therefore may be dif¢cult to recognize. Knowledge of the 3D seismic acquisition geometry and parameters used in the processing may
help the interpreter to detect potential artefacts. Critical
examination of the seismic data is essential before interpretation starts and the way the data are displayed can be
critical with respect to identi¢cation of potential artefacts.
Knowledge of the artefact described in this paper may
be of importance to those working with both seismic
processing and seismic interpretation.
ACKNOWLEDGEMENTS
The European Communities project TriTex (IST-199920500) and Norsk Hydro ASA are acknowledged for funding the research project. Norsk Hydro ASA, Statoil ASA,
Eni Norge A/S, Fortum Petroleum A/S and former Saga
Petroleum ASA are acknowledged for providing the seismic and well data.We o¡er our sincere thanks to A. Andre-
sen, G. S. Boulton, G. Elvebakk, E. Lebesbye, M. Midtb,
S. A. S. Pedersen and C. J. Pudsey for valuable input and
fruitful discussions on an earlier version of the manuscript. C. M. degrd and J. P. Holm are acknowledged
for their assistance with ¢gures and L.Wilson for correcting the English language.The reviewers J. Bulat, M. Huuse
and P. Brabham are acknowledged for insightful feedback.
We thank S. Buenz and S. Schwenke for help with the maps
in Figs 1 and 2 and for keeping the Geoquest software running. Seismic interpretation was carried out on a UNIX
seismic workstation, using the GeoFrame software interpretation package. The University of Troms acknowledges GeoQuest for computer software and guidance on
technical issues. This manuscript is also a contribution
to the Strategic University Programme ‘Sedimentary
Processes and Palaeoenvironment on the Northern Continental Margins (SPONCOM)’ funded by the Research
Council of Norway.
REFERENCES
Aber, J.S., Croot, D.G. & Fenton, M.M. (1989) Glaciotectonic
Landforms and Structures. Kluwer Academic Publishers, Dordrecht 201pp.
Andersen, L.T., Hansen, D.L. & Huuse, M. (2005) Numerical
modelling of thrust structures in unconsolidated sediments:
implications for glaciotectonic deformation. J. Struct. Geol.,
27, 587^596.
Andreassen, K., Laberg, J.S. & Vorren, T.O. (in press) Seismic sea£oor geomorphology of the Barents Sea and its glacidynamic implications. Accepted in Geomorphology.
Andreassen, K., Nilssen, L.C., Rafaelsen, B. & Kuilman,
L.W. (2004) Three-dimensional seismic data from the Barents
Sea margin reveal evidence of past ice streams and their
dynamics. Geology, 38, 729^732.
Andreassen, K., degaard, C. & Rafaelsen, B. (2007)
Imprints of Former Ice Streams, Imaged and Interpreted
Using Industry 3D Seismic Data from the South Western
Barents Sea. In: Seismic Geomorphology: Applications to
Hydrocarbon Exploration and Production (Ed. by R. Davies, H.W.
Posamentier, L.J. Wood & J.A. Cartwright). Geol. Soc. Lond.
Spec. Pub., 277, 151^169.
Astakhov,V. (2001) The stratigraphic framework for the Upper
Pleistocene of the glaciated Russian Arctic: changing paradigms. Global Planet. Change, 31, 283^295.
Boulton, G.S.,van der Meer, J.J.M., Beets, D.J., Hart, J.K. &
Ruegg, G.H.J. (1999) The sedimentary and structural evolution of a recent moraine complex, Holmstrmbreen, Spitsbergen. Quater. Sci. Rev., 18, 339^371.
Brown, A.R. (2005) Interpretation of Three-Dimensional Seismic
Data, 42, 6. AAPG Memoir, Tulsa, Oklahoma 541pp.
Bulat, J. (2005) Some considerations on the interpretation of
seabed images based on commercial 3D seismic in the FaroeShetland Channel. Basin Res., 17, 21^42.
Butt, F.A., Elverhi, A., Solheim, A. & Forsberg, C.F.
(2000) Deciphering late Cenozoic development of the western
Svalbard margin from ODP site 986 results. Mar. Geol., 169,
373^390.
Cartwright, J. & Huuse, M. (2005) 3D seismic technology: the
geological ‘Hubble’. Basin Res., 17, 1^20.
r 2007 The Authors. Journal compilation r 2007 Blackwell Publishing Ltd, Basin Research, 19, 87^103
101
B. Rafaelsen et al.
Clayton, L., Moran, S.R. & Bluemle, J.P. (1980) Explanatory
text to accompany the geologic map of Dakota. North Dakota
Geological Survey, Report of Investigation No. 69.
Denton, G.H. & Hughes, T.J. (1981) The Arctic ice sheet:
an outrageous hypothesis. In: The Last Great Ice Sheets
(Ed. By G.H. Denton & T.J. Hughes), pp. 437^467. Wiley,
New York.
Dowdeswell, J.A. & Siegert, M.J. (1999) Ice- sheet numerical
modelling and marine geophysical measurements of glacierderived sedimentation on the Eurasian Arctic continental
margins. Geol. Soc. Am. Bull., 111, 1080^1097.
Eidvin, T., Goll, R.M., Grogan, P., Smelror, M. & Ulleberg, K. (1998) The Pleistocene to middle Eocene stratigraphy and geological evolution of the western Barents Sea
continental margin at well site 7316/5-1 (Bjrnya West area).
Norsk GeoogiskTidsskrift, 78, 99^123.
Eidvin, T., Jansen, E. & Riis, F. (1993) Chronology of Tertiary
fan deposits o¡ the western Barents Sea: implications for the
uplift and erosion history of the Barents Shelf. Mar. Geol., 112,
109^131.
Elvebakk, G., Hunt, D.W. & Stemmerik, L. (2002) From iso lated buildups to buildup mosaics: 3D seismic shed new light
on the upper Carboniferous^Permian fault controlled carbo nate buildups, Norwegian Barents Sea. Sediment. Geol., 152, 7^
17.
Elverhi, A., Fjeldskaar, W., Solheim, A., Nyland-Berg,
M. & Russwurm, L. (1993) The Barents Sea ice sheet ^ a model of its growth and decay during the last ice maximum. Quater.
Sci. Rev., 12, 863^873.
Elverhi, A., Svendsen, J.-I., Solheim, A., Milliman, J.D.,
Mangerud, J. & Hooke, R.L. (1995) Late Quarternary sediment yield from the high Arctic Svalbard area. J. Geol., 103, 1^
17.
Evans, J., Pudsey, C., O’Cofaigh, C., Morris, P. & Domack,
E. (2005) Late Quaternary glacial history, £ow dynamics and
sedimentation along the eastern margin of the Antarctic Peninsula ice sheet. Quater. Sci. Rev., 24, 741^774.
Faleide, J.I., Solheim, A., Fiedler, A., Hjelstuen, B.O., Andersen, E.S. & Vanneste, K. (1996) Late Cenozoic evolution
of the western Barents Sea ^ Svalbard continental margin.
Global Planet. Change, 12, 53^74.
Fiedler, A. & Faleide, J.I. (1996) Cenozoic sedimentation along
the southwestern Barents Sea margin in relation to uplift and
erosion of the shelf. Global Planet. Change, 12, 75^93.
Forsberg, C.F., Solheim, A., Elverhoi, A., Jansen, E.,
Channell, J.E.T. & Andersen, E.S. (1999) The depositional
environment of the western Svalbard margin during the late
Pliocene and the Pleistocene: sedimentary facies changes at
site 986. In: Proceedings of the Ocean Drilling Program, Scienti¢c Results, 162 (Ed. by M.E. Raymo, E. Jansen, P. Blum & T.D. Herbert). ODP, Texas A & M University, College station, TX:
233^246.
Gataullin,V., Mangerud, J. & Svendsen, J.I. (2001) The extent of the late Weichselian ice sheet in the southeastern
Barents Sea. Global Planet. Change, 31, 453^474.
Gataullin, V. & Polyak, L. (1997) Glaciotectonic features,
southeastern Barents Sea. In: Glaciated Continental Margins:
An Atlas of Acoustic Images (Ed. by T.A. Davies, T. Bell, A.K
Cooper, H. Josenhans, L. Polyak, A. Solheim, M.S. Stoker &
J.A. Stravers), pp. 70^71. Chapman & Hall, London.
Gataullin,V., Polyak, L., Epstein, O. & Romanyuk, B. (1993)
Glacigenic deposits of the Central Deep: a key to the Late
102
Quaternary evolution of the eastern Barents Sea. Boreas, 22,
47^58.
Hambrey, M.J., Huddart, D., Bennett, M.R. & Glasser, N.F.
(1997) Genesis of ‘hummocky moraines’ by thrusting in glacier ice: evidence from Svalbard and Britain. J. Geol. Soc. Lond.,
154, 623^632.
Huuse, M. & Lykke-Andersen, H. (2000) Large- scale glacio tectonic thrust structures in the eastern Danish North Sea. In:
Deformation of Glacial Materials (Ed. by A.J. Maltman, B. Hubbard & M.J. Hambrey), Geol. Soc. Spec. Publ., 176, 293^305.
Klint, K.E.S. & Pedersen, S.A.S. (1995) The Hanklit Glacio tectonic Thrust Fault Complex, Mors, Denmark. Geol. Survey
Denmark, 35, 1^30.
Kuvaas, B. & Kristoffersen,Y. (1996) Mass movements in glaciomarine sediments on the Barents Sea continental slope.
Global Planet. Change, 12, 287^307.
Larsen, E., Kjr, K.H., Jensen, M., Demidov, I.N, Hkansson, L. & Paus, A. (2006) Early Weichselian palaeoenvironments reconstructed from a mega- scale thrust-fault complex,
Kanin Peninsula, northwestern Russia. Boreas, 35, 476^492.
Lebesbye, E. (2000) Late Cenozoic glacial history of the southwestern Barents Sea. PhD Thesis, Troms, University of
Troms, 172pp.
O’Cofaigh, C., Dowdeswell, J.A., Aller, C.S., Hiemstra,
J.F., Pudsey, C.J., Evans, J. & Evans, D.J.A. (2005) Flow dynamics and till genesis associated with a marine-based Antarctic palaeo -ice stream. Quater. Sci. Rev., 24, 709^740.
Pedersen, S.A.S. (1987) Comparative studies of gravity tectonics
in Quaternary sediments and sedimentary rocks related to fold
belts. In: Deformation of Sediments and Sedimentary Rocks (Ed. by
M.E. Jones & R.M.F Preston), Geol. Soc. Spec. Publ. 29, 165^180.
Pedersen, S.A.S. (1996) Progressive glaciotectonic deformation
in Weichselian and Palaeogene deposits at Feggeklit, northern
Denmark. Bull. Geol. Soc. Denmark, 42, 153^174.
Pedersen, S.A.S (2000) Superimposed deformation in glacio tectonics. Bull. Geol. Soc. Denmark, 46, 125^144.
Rafaelsen, B., Andreassen, K., Kuilman, L.W., Lebesbye,
E., Hogstad, K. & Midtb, M. (2002) Geomorphology of
buried glacigenic horizons in the Barents Sea from 3-dimensional seismic data. In: Glacier-In£uenced Sedimentation on HighLatitude Continental Margins (Ed. by J.A. Dowdeswell & C.
O’Cofaigh), Geol. Soc. Lond. Spec. Publ. 203, 259^276.
Rafaelsen, B., Andreassen, K., Samuelsberg, T.J., Hog stad, K. & Randen, T. (2003) Upper Paleozoic carbonate
build-ups in the Norwegian Barents Sea: new insights from
3-D seismic and automated facies mapping. 65th EAGE Conference & Exhibition, Stavanger. 2^5. June. Extended abstract,
4pp.
Siegert, M.J. & Dowdeswell, J.A. (2002) Late Weichselian iceberg, meltwater and sediment production from the Eurasian
High Arctic ice sheet: results from numerical ice- sheet modelling. Mar. Geol., 188, 109^127.
Solheim, A. & Kristoffersen, Y. (1984) Sediments above the
upper regional unconformity: thickness, seismic stratigraphy
and outline of the glacial history. Norsk Polarinstitutt Skrifter,
179B, 26pp.
Solheim, A., Russwurm, L., Elverhi, A. & Berg, M.N. (1990)
Glacial geomorphic features in the northern Barents Sea: direct
evidence for grounded ice and implications for the pattern of
deglaciation and late glacial sedimentation. In: Glacimarine
Environments: Processes and Sediments (Ed. by J.A. Dowdeswell &
J.D. Scourse), Geol. Soc. Spec. Publ., 53, 253^268.
r 2007 The Authors. Journal compilation r 2007 Blackwell Publishing Ltd, Basin Research, 19, 87^103
Glaciotectonic thrust sheets: facts or artefacts?
Stokes, C.R. & Clark, C.D. (2002) Are long subglacial bedforms indicative of fast ice £ow? Boreas, 31, 239^249.
Sttem, J. (1990) Glaciotectonic forms and structures on the
Norwegian continental shelf: observations, processes and implications. Norsk GeologiskTidsskrift, 70, 81^94.
Sttem, J. (1994) Glaciotectonic structures along the Barents
shelf margin. In: Formation and Deformation of Glacial Deposits
(Ed. by W.P. Warren & D.G. Croot), pp. 95^113. Balkerna, Rotterdam, the Netherlands.
Sttem, J., Rise, L. & Westgaard, D.A. (1992) Composition
and properties of Glacigenic sediments in the south-western
Barents Sea. Mar. Geotechnol., 10, 229^255.
Van der Wateren, F.M. (1995) Structural geology and sedimentology of push moraines. Mededelingen Rijks Geologische Dienst,
54 168pp.
Vorren, T.O., Kristoffersen, Y. & Andreassen, K. (1986)
Geology of the inner shelf west of North Cape, Norway. Norsk
Geol.Tidskrift, 66, 99^105.
Vorren,T.O. & Laberg, J.S. (1996) Late glacial air temperature,
oceanographic and ice sheet interactions in the southern
Barents Sea region. In: Late Quaternary Palaeoceanography
of the North Atlantic Margins (Ed. by J.T. Andrews, W.E.N. Austin, H. Bergsten & A.E. Jennings), Geol. Soc. Spec. Publ., 111,
303^321.
Vorren,T.O. & Laberg, J.S. (1997) Trough mouth fans ^ palaeo climate and ice- sheet monitors. Quarter. Sci. Rev., 16,
865^881.
Vorren, T.O., Lebesbye, E. & Larsen, K.B. (1990) Geometry
and genesis of the glacigenic sediments in the southern
Barents Sea. In: Glacimarine Environments: Processes and
Sediments (Ed. by J.A. Dowdeswell & J.D. Scourse), Geol. Soc.
Spec. Publ., 53, 269^288.
Vorren, T.O., Richardsen, G., Knutsen, S.-M. &
Henriksen, E. (1991) Cenozioc erosion and sedimentation
in the western Barents Sea. Mar. Petrol. Geol., 8(3), 317^340.
Manuscript received 27 February 2006; Manuscript accepted
11January 2007.
r 2007 The Authors. Journal compilation r 2007 Blackwell Publishing Ltd, Basin Research, 19, 87^103
103