Quaternary Science Reviews xxx (2012) 1e10
Contents lists available at SciVerse ScienceDirect
Quaternary Science Reviews
journal homepage: www.elsevier.com/locate/quascirev
Ice sheet retreat dynamics inferred from glacial morphology of the central
Pine Island Bay Trough, West Antarctica
Martin Jakobsson a, *, John B. Anderson b, Frank O. Nitsche c, Richard Gyllencreutz a, Alexandra E. Kirshner b,
Nina Kirchner d, Matthew O’Regan e, Rezwan Mohammad a, Björn Eriksson a
a
Department of Geological Sciences, Stockholm University, 106 91 Stockholm, Sweden
Department of Earth Sciences, Rice University, Houston, TX 77005, USA
Lamont-Doherty Earth Observatory of Columbia University, Route 9 W, Palisades, NY 10964, USA
d
Department of Physical Geography and Quaternary Geology, Stockholm University, 106 91 Stockholm, Sweden
e
School of Earth and Ocean Sciences, Cardiff University, Wales, United Kingdom
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 26 September 2011
Received in revised form
23 December 2011
Accepted 27 December 2011
Available online xxx
Pine Island Glacier drains portions of the West Antarctic Ice Sheet into the Amundsen Sea. During the Last
Glacial Maximum the glacier extended nearly 500 km from its present location onto the outer continental
shelf. Unusually restricted sea-ice cover during the austral summer of 2010 allowed for a systematic
multibeam swath-bathymetric and chirp sonar survey of the mid-shelf section of Pine Island Trough. The
mapped glacial landforms reveal new information about the paleo-Pine Island Ice Stream’s dynamic retreat
from the mid-shelf area and confirm previous suggestion of a retreat in distinct steps. The periods of
grounding line stability during the overall retreat phase are marked by sediment accumulations, i.e.
grounding zone wedges. These wedges are here mapped in sufficient detail to characterize spatial
dimensions and estimate the volume of deposited sediment. Considering a range of sediment flux rates
from the paleo-Pine Island Ice Stream we estimate that the largest and most clearly defined grounding
zone wedge, located at about 73 S in the surveyed area, took between 600 and 2000 years to form. The ice
stream retreated landward of this wedge before 12.3 cal ka BP. The swath-bathymetric imagery of landforms in Pine Island Trough includes glacial features that suggest that retreat between periods of
grounding line stability may be associated with episodes of ice shelf break-up. The depths of grounding line
wedges decrease in a landward direction, from 740 to 670 m, and record elevation of the grounding line as
it stepped landward. In all, the grounding line elevation varied by only w80 m over a distance of just over
100 km, implying a low ice sheet profile during retreat. Finally, we revisited seismic reflection profile
NB9902, acquired along Pine Island Trough in 1999, in combination with the newly acquired swathbathymetric imagery from 2010. Together these data show that the ice stream paused during its retreat to
form grounding zone wedges at an area in central Pine Island Trough where a high in dipping bedrock
strata exists and the glacial trough is narrow, forming a bathymetric “bottle neck”.
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
West Antarctica
Ice streams
Grounding zone wedge
Multibeam
1. Introduction
The West Antarctic Ice Sheet (WAIS) comprises approximately
10% of the entire Antarctic Ice Sheet volume (Lythe et al., 2001). It is an
ice sheet with its base mainly grounded below sea level, making it
particularly sensitive to environmental change and potentially
unstable. Sea level would rise more than 3 m if the marine-based part
of WAIS disintegrated (Bamber et al., 2009). Slightly less than two
thirds of WAIS drains into the Ross and RonneeFilchner Ice shelves
* Corresponding author. Tel.: þ46 8 164719.
E-mail address: [email protected] (M. Jakobsson).
while the remaining part drains into the Amundsen Sea through Pine
Island and Thwaites glaciers (Fig. 1) (Rignot et al., 2008). These latter
two glaciers were considered by Hughes (1981) as the WAIS “weak
underbelly”, referring to their direct exposure to the ocean and lack of
protective physiographic barriers or large buttressing ice shelves. In
addition, the bedrock of these glaciers deepens significantly inland,
adding to their vulnerability (Holt et al., 2006; Vaughan et al., 2006).
Recent observations indeed show rapid thinning and near constant
acceleration of the Pine Island Glacier (PIG) (e.g. Joughin et al., 2003).
The thinning is suggested to be caused by intrusion of warmer ocean
water beneath the relatively small ice shelf in front of the grounded
glacier (Jenkins et al., 2010; Jacobs et al., 2011). Inflow of warmer
water has recently also been documented west of Pine Island Bay
0277-3791/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.quascirev.2011.12.017
Please cite this article in press as: Jakobsson, M., et al., Ice sheet retreat dynamics inferred from glacial morphology of the central Pine Island Bay
Trough, West Antarctica, Quaternary Science Reviews (2012), doi:10.1016/j.quascirev.2011.12.017
2
M. Jakobsson et al. / Quaternary Science Reviews xxx (2012) 1e10
Fig. 1. Overview map showing tracklines of IB Oden during expedition OSO0910 (black lines), the general bathymetry of Pine Island Bay continental shelf (Nitsche et al., 2007) and
locations of grounding zone wedges GZW1-5. Red dots show locations of sound velocity profiles (CTD, XCTD, XBT, and XSV, see text) used to calibrate the multibeam system. The
yellow line show chirp profile PI-125A and the orange line seismic profile NBP9902, both shown in Fig. 6. PIT ¼ Pine Island Trough; PITE ¼ Pine Island Trough East; PITW ¼ Pine
Island Trough West. The black box outlines the detailed map of acquired multibeam data shown in Fig. 2a. (For interpretation of the references to colour in this figure legend, the
reader is referred to the web version of this article.)
(PIB) in the glacial trough offshore of western PIB and Getz and
Dotson ice shelves (Walker et al., 2007; Wåhlin et al., 2010). The
current behavior of the PIG highlights the need to better understand
its grounding line retreat history; how fast has it retreated since the
Last Glacial Maximum (LGM) and what are the main drivers of rapid
retreat?
There is now compelling geomorphic and chronostratigraphic
evidence for the WAIS having extended across the outer continental
shelf during the LGM (Anderson et al., 2002; Lowe and Anderson,
2002; Ó Cofaigh et al., 2002; Evans et al., 2006; Hillenbrand et al.,
2009; Mosola and Anderson, 2006; Smith et al., 2011). In the
Amundsen Sea, the present shelf bathymetry is dominated by crossshelf glacial troughs (Evans et al., 2006; Nitsche et al., 2007) marking
the paths of former ice streams that advanced across the shelf during
past glacial maxima (Lowe and Anderson, 2002; Graham et al.,
2010b). One of these troughs is the Pine IslandeThwaites Trough
(PIT) (Fig. 1). Here, the Pine Island ice stream extended to the continental shelf edge and delivered water-saturated sediments generating turbidity currents that formed gully-channel systems on the
uppermost continental slope (Dowdeswell et al., 2006a). This implies
that the Pine Island Glacier extended more than 500 km from its
present position during LGM. The central part of PIT has been sparsely
mapped due to persistent sea-ice cover (Nitsche et al., 2007),
although sets of grounding zone wedges were identified previously,
marking still-stands in the ice retreat from the LGM position (Lowe
and Anderson, 2002; Graham et al., 2010a) (Fig. 1). Prior to now,
few constraints on the dynamics of the retreat history existed. Based
on 14C-dates of calcareous foraminifera retrieved from sediment
cores, Lowe and Anderson (2002) suggested that the ice retreated
from the outer shelf before 16 ka (14C yr), but halted at about the
latitude of Burke Island where it remained long enough to produce
a massive grounding zone wedge (Fig. 1).
The Oden-Southern-Ocean 2009/2010 (OSO0910) expedition
with Swedish icebreaker IB Oden carried out swath-bathymetric
mapping, chirp sonar profiling, oceanographic station work and
coring in Pine Island Trough. Since the trough area was virtually ice
free in the austral summer of 2010, the swath-bathymetric
mapping and chirp sonar profiling were conducted as a systematic survey covering 4140 km2 of the mid-shelf section of the PIT
(Fig. 1). Here we present the detailed bathymetric data and chirp
sonar profiles along with a previously acquired seismic reflection
profile that provide new insights into the glacial dynamics during
the deglaciation of the continental shelf. The landforms imaged in
PIT indicate stepwise retreat of Pine Island Ice Stream from the midshelf that was punctuated by periods of grounding line stability and
minor advance followed by episodes of rapid retreat. This is similar
to episodic retreat of ice streams in a number of Antarctic and Arctic
cross-shelf troughs (Dowdeswell et al., 2008). Complementary to
the geomorphic results presented here, Kirshner et al. (submitted
for publication) provide new age constraints and sedimentary
evidence for the changing glacial setting during grounding line
retreat. These independent data sets support episodic post-LGM
grounding line retreat within central PIT between 16.4 and
12.3 k cal yr BP.
Please cite this article in press as: Jakobsson, M., et al., Ice sheet retreat dynamics inferred from glacial morphology of the central Pine Island Bay
Trough, West Antarctica, Quaternary Science Reviews (2012), doi:10.1016/j.quascirev.2011.12.017
M. Jakobsson et al. / Quaternary Science Reviews xxx (2012) 1e10
2. Materials and methods
2.1. Multibeam swath-bathymetric mapping
The swath-bathymetric data presented in this work were
collected using a hull-mounted Kongsberg 12 kHz EM122 1 1
multibeam echo sounder. This multibeam system includes a Seatex
Seapath 200 unit for integration of GPS navigation, heading and
attitude information. The motion sensor consists of Seatex’s MRU5.
EM122 is capable of producing 432 beams per ping and apply multipinging in the intermediate water depths. This implies that the
transmit fan is duplicated after a small offset in along track tilt
resulting in a doubling of ping density. Due to the ice protection of
the EM122 transceivers, the useable across-track angular coverage
is reduced to less than 65 65 resulting in a swath-width of
typically threeefour times the water depth. The favorable ice
conditions during our survey allowed swath mapping of the central
part of PIT along a regular survey track (Fig. 1). This has not previously been possible due to severe sea-ice conditions in this part of
the bay (e.g. Evans et al., 2006; Graham et al., 2010a). Line density
was maintained with a swath overlap between about 30 and 100%.
Sound velocity control was achieved through regular CTD
(Conductivity, Temperature, Depth) stations supplemented with
XCTD (Expendable CTD), XBT (Expendable Bathy Thermograph) and
XSV (Expendable Sound Velocimeter) casts. During the 12 day-long
survey in PIT, 46 stations with sound velocity control were acquired.
All data were processed using the software Fledermaus and gridded
to a horizontal resolution of 20 20 m. Seafloor morphology was
interpreted in the 3D environment of Fledermaus and maps were
subsequently produced in the GIS software Geomedia Professional.
2.2. Chirp sonar sub-bottom profiling
The IB Oden sub-bottom profiler is a SBP 120 3 3 chirp sonar
integrated with the multibeam through the use of the same receiving
array and the capability of using the multibeam center beam depth to
automatically adjust the acquisition window. During the OSO0910
survey the chirp sonar was operated continuously using a 2.5e7 kHz
and 35 ms long pulse. The ping interval was set to a fixed rate of
500 ms in order to facilitate post processing of acoustic water column
data acquired with the EM122 for physical oceanographic studies.
Further information about the chirp sonar settings is found in the
OSO0010 cruise report (Anderson and Jakobsson, 2010).
2.3. Seismic reflection profiling
The seismic reflection profile used in this study was collected
during a 1999 austral summer cruise (NBP9902) of the RV Nathaniel B.
Palmer. Parts of the profile were published by Lowe and Anderson
(2002, 2003). A 210 in3 GI air gun was used as a seismic source and
a single-channel streamer was used as a receiver. The data were
recorded with an Elics seismic acquisition system and band-pass
filtered. The vertical resolution of the seismic profile is on the order
of 10 m. Henceforth, the full seismic profile is referred to as Line
NBP9902 (Fig. 1).
3
mode of formation and significance to ice dynamics during recession.
Fig. 3 shows all identified landforms and their relationship to each
other. These relationships will be further discussed after the individual landforms are described.
3.1. Grounding zone wedges
Sedimentary wedges, or grounding zone wedges (GZW), that
occur within troughs on the Antarctic continental shelf are interpreted to represent grounding line deposits formed during pauses in
ice sheet’s retreat from the continental shelf (Bart and Anderson,
1996; Anderson, 1999; Mosola and Anderson, 2006; Dowdeswell
et al., 2008). Their occurrence within troughs that typically occur
seaward of modern ice streams and association with mega-scale
glacial lineations and “soft” water saturated till indicates that they
are the products of deposition by fast-flowing ice streams (Alley et al.,
1989). Larger wedges are commonly 50e100 m thick and considered
to be formed by a broad range of processes, including sub-glacial
sedimentation and deposition via sediment gravity flows debouching from the grounding line (Anderson, 1999; Christoffersen et al.,
2010).
The occurrence of two GZWs within the central PIT was originally
demonstrated by using swath bathymetry data and seismic reflection
profiles (Lowe and Anderson, 2003). Subsequently, Graham et al.
(2010a) constrained the location of the front of the larger GZW
using data from two nearby multibeam swaths and referred to it as
GZW5 (Fig. 1). In total, they identified five GZWs (GZW5-1; Fig. 1).
GZW1 and 2 are located on the outer shelf and GZW3-5 within the
central trough (Fig. 1). We will use the same nomenclature when
further describing the GZWs that fall within our survey area.
The OSO0910 survey includes the full extent of GZW3, 4 and 5
(Figs. 2e4). The data show that the seafloor morphology at GZW3 is
actually comprised of a series of back-stepping transverse ridges that
deepen in a landward direction (Fig. 5). These ridges were in part
mapped by Graham et al. (2010a,b). However, it is difficult to exclude
the possibility that these transverse ridges overly a small and
morphologically subdued GZW. Together, the transverse ridges and
grounding zone wedges record the changing elevation of the
grounding line in a landward direction. The transverse ridges occur
in water depths ranging from 720 to 740 m. GZW4 is located in
705e726 m and GZW5 in 726 to 670 m water depths. We further
note that GZW5 appears to partly overly GZW4, the two wedges
together spanning about 38 km along the axis of the trough
(Fig. 4). GZW4 has a more subdued bathymetric expression relative
to GZW5, with a maximum thickness of approximately 20 m (Fig. 4).
GZW5 extends approximately 22 km along the trough axis, 12 km
across the trough and has a maximum thickness of about 56 m when
assuming its base at a water depth of 726 m (Fig. 4). However, the
wedge width across the PIT is difficult to pinpoint as its outer edges,
along the shallow sides of the trough, have been eroded by icebergs.
The inferred width uses the outer trough walls as limits since it is
evident that the wedge extends over the entire trough. The sediment
volume of GZW5 within these horizontal constraints, calculated
using the multibeam data, is approximately 12 km3.
3.2. Streamlined sub-glacial landforms
3. Results
Several types of sedimentary landforms were produced when the
ice sheet was grounded in PIT during the LGM and the subsequent
retreat. Because sedimentation rates have been very low since icesheet retreat, these landforms are expressed in detail in the
seafloor morphology (Fig. 2). The relatively high resolution of our
multibeam swath bathymetry and pristine condition of these
features allows us to describe them in detail and to interpret their
The surveyed part of PIT contains streamlined ridge-groove
features that conform to the morphological description of megascale glacial lineations (MSGL) (Clark, 1993; Clark et al., 2003;
Wellner et al., 2006). Distinct sets of MSGLs occur north and south of
GZWs 4 and 5 (Fig. 2cee). The landward set trends about 350 ,
parallel to the trough axis, while the seaward set trends 340e345
and oblique to the trough axis (Fig. 3). Cross-cutting relations show
that the seaward set is older than the landward set. Individual MSGLs
Please cite this article in press as: Jakobsson, M., et al., Ice sheet retreat dynamics inferred from glacial morphology of the central Pine Island Bay
Trough, West Antarctica, Quaternary Science Reviews (2012), doi:10.1016/j.quascirev.2011.12.017
4
M. Jakobsson et al. / Quaternary Science Reviews xxx (2012) 1e10
Fig. 2. Main geomorphic features of the PIB central trough. a) Iceberg plow ridges; b) Corrugation ridges that overprint large linear to curvilinear furrows; c) Isolated corrugation
ridges and MSGLs superimposed on GZW4; d) GZW5 overprinted by iceberg plowmarks; e) MSGLs and corrugation ridges in the southern part of central PIT landward of GZW5.
can be followed for more than 15 km in both areas and range from
>500 to 150 m in ridge-to-ridge width and have ridge heights of
about 2 to >6 m. North of GZW5, the MSGLs are overprinted by
iceberg furrows at a water depth above approximately 670 m, while
south of GZW5 iceberg furrows generally overprint MSGLs to a depth
of about 750 m.
3.3. Large curvilinear e linear furrows
North of GZW4, linear to curvilinear lineations occur in water
depths deeper (690 and 720 m) than the MSGLs and are aligned
parallel to the axis of the trough (Fig. 2b). Some furrows have more
pronounced relief with >20 m ridge height. These lineations were
Please cite this article in press as: Jakobsson, M., et al., Ice sheet retreat dynamics inferred from glacial morphology of the central Pine Island Bay
Trough, West Antarctica, Quaternary Science Reviews (2012), doi:10.1016/j.quascirev.2011.12.017
M. Jakobsson et al. / Quaternary Science Reviews xxx (2012) 1e10
5
GZW5 and shallower than w750 m south of this wedge (Fig. 2).
These features are generally not traceable for more than 2 km
before they change directions and their relief is generally <10 m.
The seabed morphology of GZW5 is also dominated by iceberg
furrows and plowmarks (Fig. 4). Deep-keeled Antarctic icebergs are
typically calved from fast-flowing ice streams and outlet glaciers
that have only limited floating margins, whereas more extensive ice
shelves often produce bergs with keels no deeper than about 300 m
(Dowdeswell and Bamber, 2007).
3.5. Ridges
One of the most conspicuous geomorphic features in the area
consists of extremely regular and small sedimentary ridges previously described by Jakobsson et al. (2011) and named corrugation
ridges. These are 1e2 m from trough to crest and separated by about
60e200 m (Fig. 2b). Within the surveyed area, corrugation ridges are
mostly constrained to the northern part of the trough (Figs. 2 and 3).
However, some isolated sets of corrugation ridges occur also within
the MSGLs (Fig. 3). The glacial dynamic connection between the large
furrows, the corrugation ridges, and the iceberg plow ridges was
interpreted by Jakobsson et al. (2011) and is further discussed below.
3.6. Transverse ridges (moraines)
The surface of GZW3 is dominated by a series of ridges with their
spatial extensions transverse to the axis of PIT (Fig. 5). They range
from a few meters to 6 m high and occur between a few hundred
meters to >1000 m apart. Some of these ridges bend and continue
along the PIT at the eastern most part of the trough, near the outer
wall (Fig. 5).
3.7. Seismic stratigraphy
previously described by Jakobsson et al. (2011) and interpreted to be
formed from an armada of grounded large icebergs discharged during
an ice shelf break-up and likely associated grounding line retreat of
the Pine Island Ice Stream. These features predate the MSGLs north of
the GZW4 as indicated from cross-cutting relationships.
Seismic profile NBP9902 extends along the deepest central part
of PIT in the southern section of the surveyed area (Figs. 1 and 6).
From the southern part of GZW5 and northward, the seismic line
continues slightly westward of the deepest part of the trough. The
seismic stratigraphy of GZW5 is thus imaged along its extension in
a slightly oblique angle. Fig. 6 displays the section between shot
point 8100 and 1600 along with chirp sonar profile PI-125A, which
is the profile most closely following NBP9902. The rest of profile
NBP9902 is shown in Supplementary data. The north-dipping strata
underlying the trough have been eroded by the ice sheet, resulting
in a prominent unconformity that is an amalgamation of several
erosion surfaces visible only on the outer shelf (Lowe and Anderson,
2002). The older strata below the unconformity dip seaward at
progressively lower angles north of GZW5 and near the shelf break
the strata are nearly horizontal (Supplementary data). Unfortunately, a lack of crossing lines makes it difficult to fully access the
inclination of the strata and the degree to which the trough has
shifted in it position on the shelf with time. GZW5 is not characterized by a specific visible acoustic layering, so it is not possible to
determine if there are several wedges stacked upon one another.
Landwards of GZW5 there is a very thin conformable unit visible
above the unconformity. The chirp profile provides limited additional information due to poor acoustic penetration. Nevertheless,
a thin (<3 ms equivalent to w2 m) conformable sediment drape can
be distinguished in many places at closer inspection (Fig. 6d) and
sediment cores confirm the existence of these surface sediments.
3.4. Iceberg plowmarks
4. Discussion and interpretation
Randomly oriented iceberg plowmarks dominate the seafloor
morphology in water depths shallower than w670 m north of
The landforms imaged by our survey provide a more detailed view
on the WAIS retreat from the central PIT than previously revealed by
Fig. 3. Interpretation of the glaciogenic geomorphic features of the central PIT and their
relationship to each another. Note that the different sets of MSGLs have different
orientations.
Please cite this article in press as: Jakobsson, M., et al., Ice sheet retreat dynamics inferred from glacial morphology of the central Pine Island Bay
Trough, West Antarctica, Quaternary Science Reviews (2012), doi:10.1016/j.quascirev.2011.12.017
6
M. Jakobsson et al. / Quaternary Science Reviews xxx (2012) 1e10
Fig. 4. 3D-view of GZWs 4 and 5. The bathymetric profile between A and B shows the wedge heights from an estimated base depth of 726 m.
the more sparse data from the region (Lowe and Anderson, 2002;
Graham et al., 2010a). Previous mapping of the outer shelf has suggested that the major PIT bifurcates into two paths, both reaching the
shelf edge (Evans et al., 2006; Dowdeswell et al., 2006a). Following
previous authors the two outer troughs of these ice stream pathways
are henceforth referred to as PITW and PITE (Fig.1). Considering these
troughs together with previously mapped landforms, three conceptual models for the Pine IslandeThwaites paleo-ice stream flow
during the LGM were presented by Graham et al. (2010a,b). Their
preferred model suggests that the different flow directions recorded
in the two troughs were formed at different times and that the ice
stream of PITW predates that of PITE (Fig. 7). Our new data do not
allow us to constrain the age of the outer shelf troughs. It is also worth
noting that the sole evidence for ice sheet grounding at the shelf
break is the occurrence of gullies on the upper slope, which are
argued to be the product of sediment-laden water flowing from
beneath the margin of the ice sheet (Anderson, 1999; Dowdeswell
et al., 2006a). However, our data do reveal a complex history of
grounding line retreat from the central part of the trough (Fig. 7). The
mapped landforms indicate that this retreat was marked by one
episode of ice shelf break-up (Jakobsson et al., 2011) while the sediment stratigraphy suggests an additional episode of ice shelf breakup (Kirshner et al., submitted for publication). Furthermore,
prolonged pauses in grounding line retreat occurred that were
associated with build-up of grounding line wedges, possibly ice shelf
regrowth, and an episode of ice stream re-advance. The relative
timing of these events is based on the depth of geomorphic features
that record grounding line depths and superposition of these
features.
The deepest part of the main trough north of GZW5-3 is oriented
towards PITE and contains corrugation ridges (Figs. 2 and 3). These
features are interpreted to have been formed during an ice shelf
Fig. 5. 3D-view showing that GZW3 overprinted by back-stepping transverse ridges (moraines) that decrease in height in a landward direction as shown in the bathymetric profile
from A to B.
Please cite this article in press as: Jakobsson, M., et al., Ice sheet retreat dynamics inferred from glacial morphology of the central Pine Island Bay
Trough, West Antarctica, Quaternary Science Reviews (2012), doi:10.1016/j.quascirev.2011.12.017
M. Jakobsson et al. / Quaternary Science Reviews xxx (2012) 1e10
7
Fig. 6. Seismic profile NBP9902 acquired in 1999 from RV Nathaniel B. Palmer (Lowe and Anderson, 2002) and chirp profile PI-125A acquired from IB Oden in 2010. The locations of
the profiles are shown in Fig. 1. Panel A) displays the section between shot points 1600 and 8100 and panel B) includes an interpretation of this section with the most prominent
reflectors marked. Reflectors indicating dipping bedrock strata are marked with red lines and the erosional unconformity is marked by a stippled purple line. The location of GZW5
is shaded in gray. Note that the seismic line extends west of the bathymetric expression of GZW4. Panel C shows chirp profile PI-125A that occurs between shot points 2000 and
6000. A zoom in of this profile is displayed in panel D. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
break-up and likely associated grounding line retreat (Jakobsson
et al., 2011). This hypothesis implies that an ice shelf fringed the
Pine Island Ice Stream while it extended across the outer-mid shelf.
We cannot exclude that this ice shelf broke up in similar fashion as
Fig. 7. Proposed LGM reconstruction for the outer Pine Island continental shelf based
on results of Graham et al. (2010a,b). Initially, an ice stream reached the shelf edge at
PITW. This ice stream was abandoned and a new one established to the east (PITE).
the Larsen A and B ice shelves in 1995 and 2002 (MacAyeal et al.,
2003), although in the case of the PIB break-up, the ice shelf must
have disintegrated all the way back to the grounding line and even
a bit beyond as the oriented features occur in water depths down to
650 m. The mechanism for the corrugation ridge formation involves
sediment being squeezed into ridges by the trailing edge of a portion
of a broken-up ice shelf rising and settling to the seafloor under tidal
influence as it drifts seaward (Jakobsson et al., 2011). Eventually, at
a pace of 60e200 m/day, the armada of large icebergs reached the
outer more shallow part of the central PIT where the icebergs
grounded to form iceberg plow ridges (Fig. 2a). It should be noted that
the grounding line retreat due to the ice shelf break-up was asymmetric and appears to have occurred only in the western part of the
trough.
The series of landward stepping transverse ridges that occur
between 720 and 740 m water depth may have been formed as the
ice-stream retreated in this eastern part of the trough following the
break-up of the ice shelf (Fig. 5). These small transverse ridges may be
similar to De Geer moraines (Lindén and Möller, 2005). Furthermore,
similar transverse ridges have elsewhere been demonstrated to be
formed annually in ice contact settings (Ottesen and Dowdeswell,
2006). If the transverse ridges mapped in this work are annual
formations, the average retreat rate ranged from hundreds of meters
to more than 1000 m per year. Eventually this retreat stopped and
a prolonged pause in grounding line retreat resulted in the formation
of GZW4 when the grounding line was located at water depths
Please cite this article in press as: Jakobsson, M., et al., Ice sheet retreat dynamics inferred from glacial morphology of the central Pine Island Bay
Trough, West Antarctica, Quaternary Science Reviews (2012), doi:10.1016/j.quascirev.2011.12.017
8
M. Jakobsson et al. / Quaternary Science Reviews xxx (2012) 1e10
between 726 and 706 m (Fig. 5). Evident from the MSGL north set is
that this still-stand was followed by an ice stream re-advance (Figs. 3
and 8b). As the grounding line was elevated to the level of GZW4 the
flow was no longer controlled by the outer trough and the ice stream
was free to follow a more westward path. Notably, the Graham et al.
(2010a,b) model would suggest that the ice stream would actually
have followed a more easterly path as it retreated, which is not
consistent with our data (Fig. 7). However, since there is no coherent
data coverage and due to increasing overprinting of iceberg scours
west of our survey area it is not possible to determine the extent of
the re-advance. The ice stream may only have advanced over GZW4
and a little distance beyond, at least as far as we mapped the MSGL
north set. From its GZW 4 location, the grounding line stepped
landward to the location of GZW5, where the depth of the grounding
line was elevated by deposition of the wedge to between 726 and
670 m (Fig. 8c).
The precise timing of the formation of a GZW is difficult to
determine as it requires that sediment units constraining both the
beginning and end of wedge formation are identified and dated. Our
radiocarbon data are not sufficient to constrain the exact timing of
GZW formation (Kirshner et al., submitted for publication). However,
the GZW50 s morphology is mapped in detail allowing us to calculate
that the wedge’s sediment volume per meter of ice-stream width is
approximately 1 106 m3, i.e. the same volume as previously
concluded per meter width of GZW5 by Graham et al. (2010a,b). The
volume per meter ice stream-width, together with estimations of
sediment flux per year of the Pine Island and Thwaites Ice Streams,
allows us to evaluate the time required to construct GZW5 and, thus,
the time of one of Pine Island Ice Stream’s still-stands during its
general retreat from the central trough. If we assume that the
transverse moraines located on GZW3 (Fig. 5) are annual retreat
features, it is possible to estimate the sediment flux of the Pine Island
ice stream when it was active at this location of PIT. The three larger
ridges R1-R3 range in volume from w1260e1620 m3 m1 while the
three smaller ones R4eR6 range from w200e300 m3 m1, calculated
for a 1 m wide section along profile AeB shown in Fig. 5. Thus, these
ridges suggest yearly sediment fluxes of between 200 and
1620 m3 a1 m1 (meter ice stream width).
The recent study of the Kamb Ice Stream in Ross embayment
suggests that sediment wedges near grounding lines in that area
were formed mostly by melt out of basal debris (Christoffersen
et al., 2010). Assuming different scenarios for Kamb Ice Stream’s
active and stagnant phases over time and ice flow speeds between
500 m a1 (active) and 0 m a1 (stagnant), Christoffersen et al.
(2010) calculated average sediment fluxes between 525 and
875 m3 a1 m1. This is within range of the sediment fluxes of about
w100e800 m3 a1 m1 estimated for the Marguerite Bay paleo-ice
stream, which flowed across the western continental shelf of the
Antarctic Peninsula (Dowdeswell et al., 2004). Considering that the
mapped landforms in PIT all seem to suggest a highly dynamic ice
stream behavior and that the geology of the outer shelf is characterized by easily eroded sedimentary strata (Anderson, 1999), low
sediment fluxes near 100 m3 a1 m1 seems unlikely. If we consider
flux rates between 500 and 1650 m3 a1 m1, it would have taken
between 600 and 2000 years to form GZW5 (Fig. 9). The Pine Island
Ice Stream retreated from the location of GZW5 before
12.3 cal ka BP, a timing inferred from dated foraminifera in a sediment core retrieved just landward of the wedge (Fig. 8d; Kirshner
et al., submitted for publication).
A major question is why the Pine Island ice stream changed path
during the re-advance that followed the ice shelf break-up,
grounding line retreat and subsequent still-stand and formation of
Fig. 8. Ice retreat from the Pine Island outer and central continental shelf based on previous work and the interpreted glaciogenic features mapped in this study. Map a) shows the
oldest reconstruction while d is the youngest when the ice finally withdraw from the area at about 12.3 cal ka BP.
Please cite this article in press as: Jakobsson, M., et al., Ice sheet retreat dynamics inferred from glacial morphology of the central Pine Island Bay
Trough, West Antarctica, Quaternary Science Reviews (2012), doi:10.1016/j.quascirev.2011.12.017
M. Jakobsson et al. / Quaternary Science Reviews xxx (2012) 1e10
9
Acknowledgment
The expedition OSO0910 was carried out as collaboration
between Swedish Polar Research Secretariat, the Swedish Research
Council and the US National Science Foundation (NSF). We thank the
Captain and Crew of the IB Oden. Financial support was received from
the Swedish Research Council (VR), the Swedish Royal Academy of
Sciences through a grant financed by the Knut and Alice Wallenberg
Foundation and the Bert Bolin Centre for Climate Research at Stockholm University. F. Nitsche was supported by NSF grant ANT-0838735
and Anderson by NSF/ARRA grant ANT-0837925. We thank one
anonymous reviewer and Alastair Graham for insightful comments
that improved the manuscript.
Appendix. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.quascirev.2011.12.017.
References
Fig. 9. Calculated formation time of GZW5 based on its volume and assuming sediment
flux rates from the Pine Island paleo-ice stream between 500 and 1650 m3 a1 m1. The
range of flux rates estimated by Dowdeswell et al. (2004) and Christoffersen et al., 2010 are
inferred as a reference (denoted D. and Ch. respectively).
GZW4? Even if the re-advance was minor, it still changed the flow
path as evident from the MSGL north set (Fig. 3).
It has been suggested that trough morphology and underlying
geology influence ice stream flow (Dowdeswell et al., 2006b). The
one available seismic profile (NBP9902) does not provide a clear
answer to whether the geology in PIB could have influenced ice
stream behavior (Fig. 6). However, the locations of GZWs 5e3, and
thus the location for grounding line stability, do coincide with a high
in the dipping strata that may indicate more resistant layers, and thus
suggests that this may constitute an area of greater bed resistance to
the ice stream flow. In addition, both GZW4 and GZW5 occur in
locations of ‘bottle-necking’ in the trough. This, plus the fact that
lineation direction occurs at a location where the trough orientation
changes, highlights the importance of trough bathymetry on ice
stream flow direction.
5. Conclusions
In conclusion, the glacial retreat from the mid-shelf section of
PIT was highly dynamic, involving both ice shelf break-up and
associated grounding line retreat. The retreat appears to have been
episodic in accordance with typical retreat behavior previously
suggested for the last deglaciation of polar continental shelves
(Dowdeswell et al., 2008). The ice retreated from the mid-shelf
section of PIT first at about 12.3 cal ka BP to a location southward
of our surveyed area. This landward grounding line location was
likely coincident with an underlying bedrock change from seaward
dipping sedimentary strata to crystalline bedrock, which has been
shown to be the location of a notable change in glacial landforms
(Lowe and Anderson, 2002, 2003). The depths of grounding line
wedges decrease in a landward direction, from 740 to 670 m, and
record elevation of the grounding line as it stepped landward. In all,
the grounding line elevation varied by only w80 m over a distance
of just over 100 km, likely implying a low ice sheet profile. The
calculated volume of GZW5 suggests that the grounding line in the
mid-shelf section of PIT was stabile, before it finally retreated from
the area, during a time period between about 600 and 2000 years
if sediment flux rates between 500 and 1650 m3 a1 m1 are
considered.
Alley, R.B., Blankenship, D.D., Rooney, S.T., Bentley, C.R., 1989. Sedimentation beneath
ice shelves e the view from ice stream B. Marine Geology 85, 101e120.
Anderson, J., Jakobsson, M., OSO0910 Shipboard Scientific Party, 2010. Oden Southern
Ocean 0910: Cruise Report, Meddelanden Från Stockholms Universitets Institution För Gelogiska Vetenskaper. Department of Geological Sciences, Stockholm,
134 pp.
Anderson, J.B., Shipp, S.S., Lowe, A.L., Wellner, J.S., Mosola, A.B., 2002. The Antarctic
ice sheet during the Last Glacial Maximum and its subsequent retreat history:
a review. Quaternary Science Reviews 21, 49e70.
Anderson, J.B., 1999. Antarctic Marine Geology. Cambridge University Press,
Cambridge.
Bamber, J.L., Riva, R.E.M., Vermeersen, B.L.A., LeBrocq, A.M., 2009. Reassessment of
the Potential sea-level rise from a collapse of the West Antarctic ice sheet.
Science 324, 901e903.
Bart, P.J., Anderson, J.B., 1996. Seismic expression of depositional sequences associated with expansion and contraction of ice sheets on the northwestern
Antarctic Peninsula continental shelf. In: De Batist, M., Jacobs, P. (Eds.), Geology
of Siliclastic Shelf Seas. Geological Society, New York, pp. 171e186.
Christoffersen, P., Tulaczyk, S., Behar, A., 2010. Basal ice sequences in Antarctic ice
stream: exposure of past hydrologic conditions and a principal mode of sediment transfer. Journal of Geophysical Research 115, F03034.
Clark, C.D., Tulaczyk, M., Stokes, C.R., Canals, M., 2003. A groove-ploughing theory
for the production of mega-scale glacial lineations, and implications for icestream mechanics. Journal of Glaciology 49, 240e256.
Clark, C.D., 1993. Mega-scale glacial lineations and cross-cutting ice-flow landforms.
Earth Surface Processes and Landforms 18, 1e29.
Dowdeswell, J.A., Bamber, J.L., 2007. Keel depths of modern Antarctic icebergs and
implications for sea-floor scouring in the geological record. Marine Geology
243, 120e131.
Dowdeswell, J.A., Ó Cofaigh, C., Pudsey, C.J., 2004. Thickness and extent of the
subglacial till layer beneath an Antarctic paleo-ice stream. Geology 32, 13e16.
Dowdeswell, J.A., Evans, J., Ó Cofaigh, C., Anderson, J.B., 2006a. Morphology and
sedimentary processes on the continental slope off Pine Island Bay,
Amundsen Sea, West Antarctica. Geological Society of America Bulletin 118,
606e619.
Dowdeswell, J.A., Ottesen, D., Rise, L., 2006b. Flow switching and large-scale
deposition by ice streams draining former ice sheets. Geology 34, 313e316.
Dowdeswell, J.A., Ottesen, D., Evans, J., Ó Cofaigh, C., Anderson, J.B., 2008.
Submarine glacial landforms and rates of ice-stream collapse. Geology 36,
819e822.
Evans, J., Dowdeswell, J.A., Ó Cofaigh, C., Benham, T.J., Anderson, J., 2006. Extent and
dynamics of the West Antarctic ice sheet on the outer continental shelf of Pine
Island bay during the last glaciation. Marine Geology 230, 53e72.
Graham, A.G.C., Larter, R.D., Gohl, K., Dowdeswell, J.A., Hillenbrand, C.-D., Smith, J.A.,
Evans, J., Kuhn, G., Deen, T., 2010a. Flow and retreat of the Late Quaternary Pine
IslandeThwaites palaeo-ice stream, West Antarctica. Journal of Geophysical
Research 115, F03025.
Graham, A.G.C., Nitsche, F.O., Larter, R.D., 2010b. An improved bathymetry compilation for the Bellingshausen Sea, Antarctica, to inform ice-sheet and ocean
models. The Cryosphere Discuss 4, 2079e2101.
Hillenbrand, C.D., Ehrmann, W., Larter, R.D., Benetti, S., Dowdeswell, J.A.,
Ó Cofaigh, C., Graham, A.G.C., Grobe, H., 2009. Clay mineral provenance of
sediments in the southern Bellingshausen Sea reveals drainage changes of the
West Antarctic ice sheet during the Late Quaternary. Marine Geology 265,
1e18.
Holt, J.W., Blankenship, D.D., Morse, D.L., Young, D.A., Peters, M.E., Kempf, S.D.,
Richter, T.G., Vaughan, D.G., Corr, H.F.J., 2006. New boundary conditions for the
Please cite this article in press as: Jakobsson, M., et al., Ice sheet retreat dynamics inferred from glacial morphology of the central Pine Island Bay
Trough, West Antarctica, Quaternary Science Reviews (2012), doi:10.1016/j.quascirev.2011.12.017
10
M. Jakobsson et al. / Quaternary Science Reviews xxx (2012) 1e10
West Antarctic ice sheet: subglacial topography of the Thwaites and Smith
glacier catchments. Geophysical Research Letters 33, L09502.
Hughes, T.J., 1981. The weak underbelly of the West Antarctic ice-Sheet. Journal of
Glaciology 27, 518e525.
Jacobs, S.S., Jenkins, A., Giulivi, C.F., Dutrieux, P., 2011. Stronger ocean circulation and
increased melting under Pine Island Glacier ice shelf. Nature Geoscience 4,
519e523.
Jakobsson, M., Anderson, J.B., Nitsche, F.O., Dowdeswell, J.A., Gyllencreutz, R.,
Kirchner, N., O’Regan, M.A., Alley, R.B., Anandakrishnan, S., Mohammad, R.,
Eriksson, B., Fernandez, R., Kirshner, A., Minzoni, R., Stolldorf, T., Majewski, W.,
2011. Geological record of ice shelf breakup and grounding line retreat, Pine
Island bay, West Antarctica. Geology 39, 691e694.
Jenkins, A., Dutrieux, P., Jacobs, S.S., McPhail, S.D., Perrett, J.R., Webb, A.T., White, D.,
2010. Observations beneath Pine Island glacier in West Antarctica and implications for its retreat. Nature Geoscience 3, 468e472.
Joughin, I., Rignot, E., Rosanova, C.E., Lucchitta, B.K., Bohlander, J., 2003. Timing of
recent accelerations of Pine Island glacier, Antarctica. Geophysical Research
Letters 30, 1706.
Kirshner, A.E., Anderson, J.B., Jakobsson, M., O’Regan, M., Majewski, W., Nitsche, F.O.,
submitted for publication. Post-LGM deglaciation in Pine Island Bay, West
Antarctica. Quaternary Science Reviews.
Lindén, M., Möller, P., 2005. Marginal formation of De Geer moraines and their
implications to the dynamics of grounding-line recession. Journal of Quaternary
Science 20, 113e133.
Lowe, A.L., Anderson, J.B., 2002. Reconstruction of the West Antarctic ice sheet in
Pine Island Bay during the Last Glacial Maximum and its subsequent retreat
history. Quaternary Science Reviews 21, 1879e1897.
Lowe, A.L., Anderson, J.B., 2003. Evidence for abundant subglacial meltwater
beneath the paleo-ice sheet in Pine Island Bay, Antarctica. Journal of Glaciology
49, 125e138.
Lythe, M.B., Vaughan, D.G., Consortium, B., 2001. BEDMAP: a new ice thickness and
subglacial topographic model of Antarctica. Journal of Geophysical Research
106, 11335e11352.
MacAyeal, D.R., Scambos, T.A., Hulbe, C.L., Fahnestock, M.A., 2003. Catastrophic iceshelf break-up by an ice-shelf-fragment-capsize mechanism. Journal of
Glaciology 49, 22e36.
Mosola, A.B., Anderson, J.B., 2006. Expansion and rapid retreat of the West Antarctic
Ice Sheet in eastern Ross Sea: possible consequence of over-extended ice
streams? Quaternary Science Reviews 25, 2177e2196.
Nitsche, F.O., Jacobs, S.S., Larter, R.D., Gohl, K., 2007. Bathymetry of the Amundsen
Sea continental shelf: implications for geology, oceanography, and glaciology.
Gechemistry Geophysics Geosystems 8, Q10009.
Ó Cofaigh, C., Pudsey, C.J., Dowdeswell, J.A., Morris, P., 2002. Evolution of subglacial
bedforms along a paleo-ice stream, Antarctic Peninsula continental shelf.
Geophysical Research Letters 29, 1e4.
Ottesen, D., Dowdeswell, J.A., 2006. Assemblages of submarine landforms produced
by tidewater glaciers in Svalbard. Journal of Geophysical Research 111.
Rignot, E., Bamber, J.L., van den Broeke, M.R., Davis, C., Li, Y., van de Berg, W.J.,
van Meijgaard, E., 2008. Recent Antarctic ice mass loss from radar interferometry and regional climate modelling. Nature Geoscience 1, 106e110.
Smith, J.A., Hillenbrand, C.-D., Kuhn, G., Larter, R.D., Graham, A.G.C., Ehrmann, W.,
Moreton, S.G., Forwick, M., 2011. Deglacial history of the West Antarctic ice
sheet in the western Amundsen Sea embayment. Quaternary Science Reviews
30, 488e505.
Vaughan, D.G., Corr, H.F.J., Ferraccioli, F., Frearson, N., O’Hare, A., Mach, D., Holt, J.W.,
Blankenship, D.D., Morse, D.L., Young, D.A., 2006. New boundary conditions for
the West Antarctic ice sheet: subglacial topography beneath Pine Island glacier.
Geophysical Research Letters 33, L09501.
Wåhlin, A.K., Yuan, X., Björk, G., Nohr, C., 2010. Inflow of Warm Circumpolar Deep water
in the Central Amundsen Shelf. Journal of Physical Oceanography 40, 1427e1433.
Walker, D.P., Brandon, M.A., Jenkins, A., Allen, J.T., Dowdeswell, J.A., Evans, J., 2007.
Oceanic heat transport onto the Amundsen Sea shelf through a submarine
glacial trough. Geophysical Research Letters 34.
Wellner, J.S., Heroy, D.C., Anderson, J.B., 2006. The death mask of the Antarctic ice
sheet: comparison of glacial geomorphic features across the continental shelf.
Geomorphology 75, 157e171.
Please cite this article in press as: Jakobsson, M., et al., Ice sheet retreat dynamics inferred from glacial morphology of the central Pine Island Bay
Trough, West Antarctica, Quaternary Science Reviews (2012), doi:10.1016/j.quascirev.2011.12.017