Post-Megaslide Slope Stability North
of Svalbard, Arctic Ocean
D. Winkelmann, W.H. Geissler, R. Stein, and F. Niessen
Abstract In the light of a warming globe, increasing coastal population and human
offshore activities, slope stability issues steadily gain significance. The Arctic Ocean
is predicted to exhibit most drastic changes. Following the enormous Hinlopen/
Yermak Megaslide north of Svalbard 30,000 years ago, the adjacent slopes developed
several failure types as a consequence of the partial removal of the Hinlopen trough
mouth fan. The local slope to the east is structured by several detachment surfaces
that facilitate large scale creeping. This soft sediment deformation includes turbulent
structures like folds on a meter-scale. The creeping sediments partly cover the eastern
main slide debris of the megaslide within Sophia Basin. The timing of this gravitydriven mass transport can roughly be assessed by the time interval that occurred
between the megaslide and today. These features mark the slope as unstable.
Keywords Submarine landslide • mass-failure • tsunami • geohazard • seafloor
morphology • submarine slope stability • submarine debris flow • post-slide
slope stability
1
1.1
Introduction
Indication for Slope Failure?
Large-scale gravity-driven slope failures (e.g. submarine landslides) represent a
natural hazard to any sea-floor infrastructure as well as to coastal communities due
to their ability to generate large-scale tsunamis. The predictability of these events
D. Winkelmann ()
Leibniz Institute for Marine Science (IFM-GEOMAR), Kiel, Germany
e-mail: dwinkelmann@ifm-geomar.de
W.H. Geissler, R. Stein, and F. Niessen
Alfred Wegener Institute for Polar and Marine Research, Bremerhaven, Germany
D.C. Mosher et al. (eds.), Submarine Mass Movements and Their Consequences,
Advances in Natural and Technological Hazards Research, Vol 28,
© Springer Science + Business Media B.V. 2010
279
280
D. Winkelmann et al.
may be inferred from recurrence rates or from numerical modelling. Both
approaches are based on the assumption that the cause of the slides and their
trigger-mechanisms are fully described. The intrinsic problem with investigations
of slope failures is that the remaining slope (sedimentary properties, structures etc.)
actually characterises a stable environment (it has not failed). Thus, the identification of indications for future failure remains a key target. Among the variety of
proposed trigger-mechanisms, one refers to the acceleration of slow failures along
detachment horizons (e.g. creeping sediment) leading to sliding.
In the light of a warming globe, increasing coastal population and human offshore activities, slope stability issues gain steadily significance. The Arctic Ocean
is predicted to exhibit most drastic changes.
1.2
Research Area
The Arctic Svalbard archipelago comprises the main islands of Spitsbergen and
Nordaustland and further islands reaching from rather large to numerous smallscale islands. It is neighboured by the semi-enclosed submarine Sophia Basin to the
north (Fig. 1). The archipelago was repeatedly glaciated by the Svalbard-Barents
Sea-Ice-Sheet (SBIS) which shaped the modern landscape with mountain ridges,
fjords and cross-shelf troughs by glacial erosion. The shelf north of Nordaustland
Fig. 1 (left) Overview map of Svalbard archipelago and the eastern Fram Strait region; (right)
close up of the Sophia Basin with the outline of the Hinlopen/Yermak Megaslide and location of
Fig. 2. Bathymetry: IBCAO2 (Jackobson et al. 2008)
Post-Megaslide Slope Stability North of Svalbard, Arctic Ocean
281
has, in comparison to the Barents Sea, hardly been investigated due to its natural
sea ice cover. The first systematic seismic surveys were carried out in 1990, 1993
(NPD-MOF-90/93) and 1999 (Geissler and Jokat 2004). The research area is the
western slope of this shelf facing the Sophia Basin. This area is situated directly
adjacent to the headwall area and deposits of a giant submarine slide.
This giant slope failure was first described by Cherkis et al. (1999) but was discussed controversially due to its enormous dimensions. Later studies (incl. two cruises
of the ESF EUROMARGINS project SPACOMA in 2004) confirmed the existence of
a submarine slide complex with thick debris deposits which occurred 30,000 years ago
(Vanneste et al. 2006; Winkelmann et al. 2006, 2008a; Winkelmann and Stein 2007).
This slide complex consists of one major failure event, the Hinlopen/Yermak
Megaslide. It removed partly the Hinlopen Trough Mouth Fan (TMF) and drastically affected the post-slide stability of the adjacent slopes leading to a variety of
slope failures (Winkelmann et al. 2008a). The potential for repeated failures, thus, is
significant. While Winkelmann et al. (2008a) focused on the megaslide’s dynamics,
we concentrate in this paper on the post-slide instability issue.
1.3
Material and Methods
Parametric echo-sounding (Atlas PARASOUND, 4 kHz parametric frequency of
18 and 22 kHz primary frequencies; footprint: 4° × 4.5°), high-resolution swath
bathymetry (Atlas HYDROSWEEP DS-2, 15.5 kHz; 118 beams, 90° insonification
angle) data were acquired aboard RV “Polarstern” during ARK-XX/3 in 2004 (Stein
et al. 2005). Additional high-resolution swath bathymetry data (Simrad EM 300,
30 kHz; 135 beams; 150° insonification angle; Vanneste et al. 2006) were integrated.
The data were gridded at 20 m resolution and adjusted for system-related offsets.
Slope angles are based on these grids. Further, we interpreted multi-channel reflection
seismic data of profile MF-908100 acquired during the regional survey NPD-MOFF-90
and owned by the Norwegian Petroleum Directorate. Gravity coring was accomplished with a giant gravity corer (35 × 35 cm; kastenlot). Core description and x-ray
radiography were performed aboard the vessel directly after recovery.
2
Results
High-resolution bathymetry reveals several bulges of sediment (frontal slope angles
>6°) on the continental slope north of the Hinlopen/Yermak Megaslide’s headwalls
(Fig. 2). The parametric echo-sounding data from a profile running down-slope
across several bulges displays staircase structures and an upper transparent unit that
drapes acoustically stratified sediments below. The acoustic profile crossing one of
these bulges more or less slope-parallel displays that the upper transparent unit is
characterised by a lateral transition into acoustical stratification (Fig. 3). This
indicates a lateral shift of the internal acoustic character of the bulge.
282
D. Winkelmann et al.
Fig. 2 Maps of the eastern Sophia Basin (see Fig. 1 for location): (left) shaded bathymetry and
(right) associated slope angles displaying creeping sediments with numerous escarpments, smallscale failures and canyons on the continental slope adjacent to the removed Hinlopen Trough
Mouth Fan. Slide deposits of the Hinlopen/Yermak Megaslide are visible in the western part of
the maps. Position of core PS66/327–3, location of seismic profile (Fig. 4) and acoustic profiles
(Fig. 3) are indicated
The kastenlot PS66/327–3 (see Fig. 2 for location) was placed into this upper
acoustically transparent unit (Fig. 3) and recovered 5.5 m of sediment. The examination of the core revealed tilted sedimentary layers involved in deformational
structures like folds at different scales. These structures are clearly visible throughout the core, identifying a heavily disturbed sediment. The uppermost 40 cm of this
core contains a fairly sorted silt unit (containing sand and gravel) with sharp contact
at its base. We interpret this uppermost 40 cm to be a turbidite in its early stage
(being less sorted). The sedimentological classification of the rest of the core would
be close to a debris flow. But since the initial structures are not completely disintegrated, it does not reflect the flow of debris in a strict sense. We interpret the
deformational structures of the core to reflect creeping of soft sediment.
3
Discussion
Textural structures in PS66/327–3 indicate soft sediment deformation. The same
range of glaciomarine sediments that were recovered within the Sophia Basin
(c.f. Winkelmann et al. 2006, 2008a, b) are obviously involved in this slope failure
type. No indication for any special failure-related sedimentary component (e.g. gaseous
sediment, high porosity sands/oozes or hydrates etc.) was found. There are several
Fig. 3 PARASOUND profiles and gravity core PS66/327–3 (5.5 × 0.35 × 0.35 m). The core recovered strongly deformed siliciclastic glaciomarine sediments
(silty clays with IRD-layers) from the upper acoustically transparent unit. For location of profiles and core see Fig. 2. Note that the lower profile changes
direction on core position (up and down the slope)
Post-Megaslide Slope Stability North of Svalbard, Arctic Ocean
283
284
D. Winkelmann et al.
conditions to produce the folding in core PS66/327–3: (a) undisturbed glaciomarine
sedimentation (with changing sedimentary input during interglacial and glacial
cycles) to produce the well developed layers; (b) deformation of these layers. Since
the cored sediments (and deformed sediments between 675 and 750 m) are below
any reasonably reconstructed grounding line of the SBIS, the remaining force
should be gravity.
Gravity-driven movement requires a sufficient inclination of the slope. Since the
sediments have been deposited quietly before their deformation, we interpret the
cause for the gravity-driven deformation to be over-steepening. The initiation of
movement may be attributed to seismic activity (repeated excitation by frequent
earthquakes) but may have another cause.
Intended to obtain a highly resolved Holocene record, the cored material
was ordered to be disposed back to sea. However, the core was photographed
and sampled for x-ray analysis before. Synoptic investigation as well as x-ray
radiography identified no significant difference in sedimentary properties of
undisturbed sediments from the Sophia Basin (c.f. Winkelmann et al. 2006,
2008a, b).
The cause for the over-steepening of the slope can be seen in the structural
adjustment of this continental slope following the removal of the Hinlopen TMF.
This western continental slope north of Nordaustland is characterised by a set of
detachments. These detachments have been speculated to be the cause for the
megaslide (Cherkis et al. 1999) and re-interpreted to simply represent boundaries
between sedimentary layers (Vanneste et al. 2006). The wavy sediments would
reflect sediment waves of contourites according to Vanneste et al. (2006). We
examined the seismic reflection profile MF-908100 (Cherkis et al. 1999) and identified several detachment surfaces that obviously facilitate large-scale deformation
within the slope. These detachments can also be seen in seismic reflection profile
04JM059 (Vanneste et al. 2006). Three main detachment surfaces are sub-horizontally
aligned and often combined with low-angle listric faults that route in these
detachments. The listric faults steepen upwards and often transform into another
detachment or low-angle fault (Fig. 4). The deepest detachment outlined by Cherkis
et al. (1999) could not be confirmed. The sediments bound between these structures
and which appear like large-scale waves are deformed packages of sediments
bound and slowly displaced between the detachments.
The upper part of the slope exhibits indications for further deformation into
the shallow sediments. The staircase structures (Fig. 3) imaged in the parametric echo-sounding data require another, shallower detachment horizon which is
not imaged in the PARASOUND data nor resolved in seismic reflection data.
We speculate that its sub-bottom depth may be between 30 and 50 m. The
creeping sediments on top obviously require another, even shallower detachment surface. This is the boundary towards the stratified sediments below
(between 2 and 7 m bsf.).
The chaotic and partly acoustically transparent signature of shallower sediments
displayed in profile 04JM059 may resemble debris flows. This may have led
Vanneste et al. (2006) to interpret these shallower slope sediments bound between
Post-Megaslide Slope Stability North of Svalbard, Arctic Ocean
285
Fig. 4 Multi-channel seismic reflection profile across the eastern continental slope adjacent to the
megaslide scar; yellow solid lines indicate position of main detachments, red solid lines indicate
low-angle or listric faults, grey solid lines inferred pre-slide and immediate post-slide sea-floor
morphology; approximated location of acoustic profiles (Fig. 3) indicated as red box in lower image
the detachments as glacigenic debris flows. However, some of those supposed
debris flows show clear stratification, some of them deformed. In fact, the cored
upper transparent unit has no clear lateral boundaries, but rather develops gently
into acoustically stratified sediments. We interpret this to be related to the systemimmanent acoustic footprint of the PARASOUND system (4 × 4.5°). The absence
of a sufficiently large and consistent (with regard to the footprint) intra-sedimentary
surfaces prevents the record of consistent internal reflectors. Thus, these sediments
appear transparent in the acoustic imagery of sedimentary structures (Fig. 4). This
indicates one problematic aspect of the acoustic identification of gravity-driven
mass movements.
The timing of this deformation is difficult to constrain since we did not recover
undisturbed sediments below and the lack of younger non-deformed sediments on
top. However, Winkelmann et al. (2008b) have shown that terrigenous input
events (TIEs) of the Sophia Basin formed discrete sedimentary layers which
develop acoustically recordable sub-buttom reflectors in the parametric echosounding data. These can be used for acoustic stratigraphy or sedimentologically
as stratigraphic marker horizons (Winkelmann et al. 2008b). Judging from the
286
D. Winkelmann et al.
PARASOUND imagery (Fig. 3), the well developed strata below the upper transparent unit resemble the sequence of TIEs of MIS 3–7 (and older below). Since
the Hinlopen/Yermak Megaslide occurred in MIS 3 (30 ky ago), this would corroborate a post-megaslide age of the deformation. If correct, this would also
confirm the gravity-driven mechanism as a result of slope over-steepening.
However, the reflectors below the deformed sediments of the upper transparent
unit may be older.
In summary, there are at least five detachments structuring the slope. The
nature of these detachments may be explained by sediment layers with low shear
strength but remains to be investigated. Their cause is probably the partial TMF
collapse that led into the Hinlopen/Yermak megaslide (Hinlopen/Yermak
Megaslide; c.f. Winkelmann et al. 2006, 2008a). The continental slope lost its
stabilisation by the adjacent TMF sediments. This led to instability, down-slope
movement along weak horizons that acted as detachment (c.f. Cherkis et al.
1999) and to creeping due to over-steepening of the upper slope. This scenario
is corroborated by toe structures. They indicate that the western part of the
megaslide’s deposits has partly been covered by slope sediments creeping
onto them. The bulges of sediment along the slide’s debris (Fig. 2) support this
interpretation.
The investigated slope appears unstable. The sediments may persistently or
repeatedly move. Whether the areas of deforming sediments are about to fail within
fast failure events (e.g. slides) remains debatable. The present detachments already
act as slip plane. Seismic excitation may stimulate creeping into faster slope failures
(slides). Younger failures in the eastern headwall area (evident as detachment
ridges; c.f. Vanneste et al. 2006; Winkelmann et al. 2008a) document the acute
failure potential.
4
Conclusion
The western shelf north of Nordaustland is structured by at least five detachments. They
have probably formed as a consequence of the partial collapse of the Hinlopen
TMF during the Hinlopen/Yermak Megaslide. The slope sediment moves downslope along these detachments and covers partly the eastern main slide debris of the
Hinlopen/Yermak Megaslide. Staircase structures and creeping sediment are identified in the upper sediments and point to over-steepening. The eastern headwalls
may likely be related to this slope instability that post-dates the megaslide. Further
high-resolution acoustic, seismic and geotechnical data are needed to understand
this soft sediment deformation.
Acknowledgments We thank captain and crew of Arctic expedition ARK-XX/3. We are thankful
to Jürgen Mienert (University of Tromsö) and Maarten Vanneste (Norwegian Geotechnical
Institute Oslo) for providing additional bathymetry data of the headwall area.
Post-Megaslide Slope Stability North of Svalbard, Arctic Ocean
287
References
Cherkis, N.Z., Max, M.D., Vogt, P.R., Crane, K., Midthassel, A., Sundvor, E., 1999, Large-scale
mass wasting on the north Spitsbergen continental margin, Arctic Ocean. Geo-Mar Lett 19:
131–142.
Jakobsson, M., Macnab, R., Mayer, L., Anderson, R., Edwards, M., Hatzky, J., Schenke, H.W.,
Johnson, P., 2008, An improved bathymetric portrayal of the Arctic Ocean: Implications for
ocean modeling and geological, geophysical and oceanographic analyses. Geophys Res Lett
35: L07602, doi:10.1029/2008GL033520.
Geissler, W.H., Jokat, W., 2004, A geophysical study of the northern Svalbard continental margin.
Geophys J Int 158: 50–66, doi: 10.1111/j.1365–246X.2004.02315.x.
Stein, R. (Ed.), 2005, Scientific Cruise Report of the Arctic Expedition ARK-XX/3 of RV
“Polarstern” in 2004: Fram Strait, Yermak Plateau and East Greenland Continental Margin,
Rep Polar Mar Res 517.
Vanneste, M., Mienert, J., Bünz, S., 2006, The Hinlopen Slide: A giant, submarine slope failure
on the northern Svalbard Margin, Arctic Ocean. Earth Planet Sci Lett 245: 373–388.
Winkelmann, D., Jokat, W., Stein, R., Winkler, A., 2006, Age and extent of the Yermak Slide north
of Spitsbergen, Arctic Ocean. Geochem Geophys Geosyst 7, Q06007, doi:10.1029/2005GC001130.
Winkelmann, D., Stein, R., 2007, Triggering of the Hinlopen/Yermak Megaslide in relation to
paleoceanography and climate history of the continental margin north of Spitsbergen.
Geochem Geophys Geosyst 8: Q06018, doi:10.1029/2006GC001485.
Winkelmann, D., Geissler, W., Schneider, J., Stein, R., 2008a, Dynamic and timing of the
Hinlopen/Yermak Megaslide north of Spitsbergen, Arctic Ocean. Mar Geol 250: 34–50.
Winkelmann, D., Schäfer, C., Stein, R., Mackensen, A., 2008b, Terrigenous events and climate
history of the Sophia Basin, Arctic Ocean. Geochem Geophys Geosyst 9: Q07023,
doi:10.1029/2008GC002038.