Accepted Manuscript Glacially influenced morphodynamic features – Examples from the north Faroe margin
Tove Nielsen, Antoon Kuijpers PII: DOI: Reference:
S0025-3227(17)30201-3 https://doi.org/10.1016/j.margeo.2018.01.007 MARGO 5749
To appear in:
Marine Geology
Received date: Revised date: Accepted date:
1 May 2017 1 September 2017 4 January 2018
Please cite this article as: Tove Nielsen, Antoon Kuijpers , Glacially influenced morphodynamic features – Examples from the north Faroe margin. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Margo(2017), https://doi.org/10.1016/j.margeo.2018.01.007
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Glacially influenced morphodynamic features – examples from the north Faroe margin Tove Nielsen1 & Antoon Kuijpers1 Geological Survey of Denmark and Greenland (GEUS), Øster Voldgade 10, 1350 Copenhagen K, Denmark
PT
1
RI
Abstract
High resolution reflection seismic data, subbottom profiler and side scan sonar information,
SC
together with results from sediment core studies, have been used to study various glacially
NU
influenced morphodynamic features on the northern Faroese continental margin. On the shelf and upper slope a thick, buried turbate complex has been found, which we estimate to have been formed
MA
between ca 470,000 and 120,000 yrs BP. We interpret this turbate to be the result of episodic, extensive iceberg grounding during extreme glaciation within the period MIS 12 – MIS 6 (Elster-
D
Saale complex). Late MIS 6 is found to be the most likely age of the last episode of turbate
PT E
formation as suggested by age/depth correlation with dated sediment cores from the nearby area. The origin of the iceberg turbate may be attributed to deglacial drift of deep-draft icebergs from
CE
icestream sources in the Arctic, East Greenland and/or northern Iceland. At the base of slope and adjacent deep-water area a complex of mud diapir features has been observed. Our data show that
AC
activation of these features are the result of density inversion processes caused by fine-grained, unconsolidated hemipelagic deposits being overlain by dense, glacigenic North Sea Fan (NSF) deposits. Due to sudden and fast accumulation of the fan sediments, a normal and regular compaction and dewatering of the underlying hemipelagic sediments could not occur; instead, subseabed sediment mobilisation took place resulting in enhanced diapir formation. A general intensification of the diapiric processes is likely related to the MIS 6 (Saalian) glacial period, as NSF-derived glacigenic debris flows reached the northern outlet of the Faroe-Shetland Channel.
1
ACCEPTED MANUSCRIPT
The most recent diapir activity is concluded to have occurred during the last (Weichselian) glaciation.
1. Introduction
PT
Relatively little information is available about the glacial history of the Faroe Islands, which in particular applies to older, pre-Weichselian glaciations and how far the various icecaps had
RI
extended offshore. It has, however, been documented that at least during the Quaternary the island
SC
group has been repeatedly covered by an icecap (Jørgensen and Rasmussen, 1986). The latest occurrence of a local icecap may have been during the Younger Dryas (Humlum et al., 1996). The
NU
marine glacial environment of the Faroe Islands has been studied by, amongst others, Nielsen et al
MA
(2007), who found that during the Weichselian glaciation, the outer shelf was likely ice-free. The authors also report the presence of an older, pre-Weichselian iceberg turbate on the northern Faroe
D
margin, presumably originating from deep-draft (> 600 m) iceberg drift in the Nordic Seas. Other
PT E
evidence of grounding of deep-draft icebergs around the Faroe Islands had previously been found on the southern flank of the Iceland-Faroe Ridge (Kuijpers and Werner, 2007). Nielsen et al. (2007) further conclude that formation of several trough systems on the eastern and western Faroe shelf
CE
may have occurred at around the same time as the iceberg turbate was formed. As a consequence of
AC
glacial sea level fall and subsequent rise during deglaciation several stages of large-scale slope instability have occurred on the Faroe margin (Kuijpers et al., 2001). Associated mass flow events could be correlated with sea level lowstand during the Last Glacial Maximum (LGM) and fast sea level rise at the Pleistocene/Holocene boundary. As mentioned above, for the Faroe Islands little information is available about older, pre-Weichselian glaciations. Instead, the glacial history of Iceland, where more details are available, may be used to also provide relevant information on the glacial history of the Faroe Islands. Using field observations, a modelling study shows that the
2
ACCEPTED MANUSCRIPT
optimum LGM ice sheet on Iceland had a mean thickness of 940 m, a plateau elevation of about 2000 m, and a substantial proportion of its base grounded below sea level (Hubbard et al., 2006). It is noteworthy, that at the same time this Icelandic LGM ice sheet was much more limited in thickness than the Saale (Marine Isotope Stage, MIS 6) ice sheet, which was extensive with a major
PT
ice stream diversion to the North (Van Vliet-Lanoë et al., 2010). Based on stratigraphic and sedimentological studies, Geirsdóttir et al (2007) report that during the last 4-5 Ma Iceland has
RI
experienced more than 20 glaciations, but that over the last 2.5 Ma glaciation became more
SC
widespread and in the past 0.5 Ma years the glacial erosion rates had significantly increased. This trend appears to correspond well with the apparent increase of Northern Hemisphere glacial ice
NU
volumes of MIS 12, 10, 6 and 4-2 (Ehlers and Gibbard, 2007).
MA
In the mid-1990s mounded seabed features were observed at the northwestern flank of the Fugloy Ridge and were initially reported as possible cold-water coral mounds (see Long et al. 2003).
D
During later studies it turned out, however, that these were mud diapirs which were informally
PT E
named the Pilot Whale Diapirs by some authors (Haflidason et al. 1996). The formation of the diapirs has been suggested as tectonically triggered upwards pushing of Eocene to Oligocene mudstone strata (Johnson et al. 2005). Alternatively, fluid-release has been proposed as a possible
CE
trigger mechanism (Long et al. 2003). These diapirs notably occur at the margin of the North Sea
AC
Fan (NSF) made of a more than 1 km thick wedge of glacigenic debris flow and slide sediments deposited since the early Pliocene and originating from the southwest Norwegian margin (King et al., 1998; Nygård et al., 2005). The main objective of the present paper is to better constrain the timing of formation of the previously reported iceberg turbate and refine the history of diapir activity at the northern Faroe margin in order to link these regional depositional processes to glacial activity in the northernmost part of the North Atlantic.
3
ACCEPTED MANUSCRIPT
2. Physiographic setting and geology of the study area The study area is part of the Norwegian Sea margin of the Faroe Platform (Fig. 1), which has a WSW-ENE trending extension, the Fugloy Ridge that separates the Norwegian Sea from the FaroeShetland channel. The regional bathymetry displays water depths ranging between around 500 m or
PT
less on the Platform and top of the Fugloy Ridge and almost 3000 m in the adjacent Norwegian Sea. The geology of this part of the Faroe margin is characterized by an up to approximately 2 km thick
RI
wedge of Cenozoic sediments overlying a basaltic basement (Nielsen and Van Weering, 1998).
SC
Nielsen and Van Weering (1998) assigned the initiation of major mass failure events and an up to
NU
300 m high slide escarpment at the middle and lower slope, i.e. the North Faroese Slide Complex (Fig. 1; Nielsen & van Weering, 1998; van Weering et al. 1998), to slope instability having
MA
occurred in the Late Pliocene. Subsequent glacial-interglacial sea level changes resulted in further slope instability and associated mass flow deposition in the study area (Kuijpers et al. 2001).
D
Furthermore, major depositional events in this area are not only related to slope instability on the
PT E
nearby Faroe margin, but also reflect episodic major sediment transport and depositional events associated with glacigenic debris flows from the more distant northern North Sea margin ( Haflidason et al., 1998; King et al., 1998; Nygård et al., 2005). The latter authors report a
CE
northwestward extension of the so-called ‘North Sea Fan’ (NSF) in the study area, after having
AC
passed the eastern extremity of the Fugloy Ridge (Fig. 1). Oceanographically, the study area is characterized by the presence of water masses that flow eastward along the northern Faroe shelf and after that southwards into the Faroe-Shetland Channel. These water masses include both cold waters from the East Icelandic Current and Norwegian Sea Deep Water mass and warmer, Atlantic water derived from the North Atlantic Current (Hansen et al., 1998; Hansen and Østerhus, 2000).
4
ACCEPTED MANUSCRIPT
3. Methods 3.1 Seismic Reflection Data High-resolution, multichannel seismic reflection data used in this study were acquired during two cruises (1991, 1993) of RV Pelagia, Royal Netherlands Institute for Sea Research (NIOZ), jointly
PT
carried out by NIOZ and GEUS (1991) and within the framework of the EU-funded ‘European North Atlantic Margin (ENAM)’ project (Van Weering et al., 1993). A cluster of four 40-in3 sleeve
RI
guns was used in combination with a 24-channel streamer. The sample interval was 1 ms and
SC
recording length 2000-3000 ms. Navigation was based on the Differential Global Positioning
NU
System (DGPS). Maximum seismic penetration is about 2 km with a vertical resolution of approx. 10 m. Interpretation was made on raw-stack data using a Landmark Workstation.
MA
Furthermore, during a ‘Training-Trough-Research’ cruise in 2012 (‘TTR-12’) with RV ‘Prof. Logachev’, St. Petersborg, (Kenyon et al. 2003) the study area was investigated using a seismic set-
D
up consisting of one 3-litre airgun and a 25 m long streamer with 50 hydrophones Shot rate was 10
PT E
s (approximately every 30 meters), recording length 3 s and sample interval 1 ms. The data was acquired digitally and subsequent processed with the RadExPro software (Kenyon et al. 2003). For
CE
investigating details of the uppermost sedimentary units, an O.R.E. hull-mounted sub-bottom profiler was operated using a transmitting frequency of 5 kHz, with a pulse length of 7ms and
AC
energy power of 2.5 kW. For the age-to-depth conversion related to age calculation of the iceberg turbate a sound velocity for the upper seabed sediment layers (silty clay with some fine sand) of around 1580 m/s was used (Hamilton and Bachman, 1981). For the water column, an acoustic velocity of 1450 m/s has been used.
3.2 Side-scan sonar
5
ACCEPTED MANUSCRIPT
During the R/V ‘Prof. Logachev’ TTR12 cruise a deep-towed, high-resolution O.R.E. MAK M1 side-scan sonar was deployed, operating at a frequency of 30 kHz, with a total swath range of 2 km (1 km per side) also including a sub-bottom profiler, operated at frequency of 5 kHz. The sonar has a variable resolution of about 7 to 1 m across-track (maximum range to centre) and along-track
PT
(centre to maximum range), respectively. Ship speed during the deep-tow operations was at a speed of 1.5-2 knots, with the tow-fish held at about 100 m above the seabed. Data processing included
RI
slant-range-to-ground-range (SLT) correction of the sonographs, geometrical correction of the
SC
profiles for recovery of the real seafloor topography, and smoothing average filtering of both types of records. Individual lines were geometrically corrected for the towing speed of the fish. Digital
NU
mosaics of the data were built, i.e. the data were plotted in real geographical co-ordinates in
MA
Mercator projection. Further side-scan data acquisition was by the OKEAN long-range side-scan sonar operated with a frequency of 9.5 kHz and a pulse length of 2 s. The swath range of the system
D
is 12 km. Receiving time was set to 14 s and the return signals were digitally recorded. The
PT E
processing and mosaic of the OKEAN data were generally the same as for the O.R.E. MAK M1 (see above). Positioning was based on both GPS and GLONASS satellite navigation, which allowed
CE
a greater accuracy.
AC
3.3 Sediment sampling and core analysis Onboard RV ‘Prof. Logachev’ a 6-m long and c. 800 kg gravity corer was used for sediment sampling. Surficial seabed sediments were collected using a Reineck Box-corer with a 0.5 m x 0.5m x 0.5 m box capable of retrieving c. 185 kg of sediments. For sediment coring onboard RV ‘Pelagia’ a 12-m long and c. 1500 kg piston corer system was deployed. Sampling positions are indicated on Figures 1 and 4.
6
ACCEPTED MANUSCRIPT
Magnetic susceptibility was measured on all the obtained gravity cores. The equipment used was a Bartington MS2E/1, designed to perform high-resolution measurements of magnetic susceptibility along flat surfaces. A sedimentological description of all the cores had previously been made and microfossil (foraminifera, diatoms, dinoflagellates) studies were performed on selected gravity core
PT
samples collected during the TTR12 cruise (Kenyon et al., 2003). For sediment dating AMS 14C measurements of selected samples from diapir cores were carried out at the P.J. Van de Graaff
RI
AMS14C Laboratory of Utrecht State University. In addition, information from previously
SC
published piston core records (Rasmussen et al., 1996; 1998; 1999; 2003) has also been used for the
NU
interpretation of the present study.
MA
4. Results 4.1 Seismic data
D
An overview of the seismic lines, acoustic profiles and sediment cores used for this study is shown
PT E
in Figure 1. Figure 2 illustrates a high-resolution seismic transect (line PE91-06) of the outer shelf and slope of the northeastern sector of the Faroe Platform. The seismic pattern of line PE91-06
CE
reveals an upper seismic unit described as Unit 9 by Nielsen and Van Weering (1998), who mapped this unit margin-wide and interpreted it as an overall contouritic deposition based on the outer shape
AC
and a dominant parallel reflection pattern (Fig. 2). However, Nielsen and van Weering also noted that the shelfward part of Unit 9 was characterized by a chaotic reflection pattern, which they interpreted as an iceberg turbate (Fig. 2). Here we focus on the transition zone that marks the described change in the internal reflection pattern of Unit 9 at around 800 m water depth (Nielsen and van Weering, 1998). Upslope of this transition zone the chaotic reflection pattern of the iceberg turbate is seen, while downslope of the zone the seismic pattern is characterised by the parallel reflections of the contourite deposits (Fig. 2). In the transition zone, the iceberg turbate is overlain
7
ACCEPTED MANUSCRIPT
by the upper part of the Unit 9 contourite deposits, here termed Unit 9a (Fig. 2). From a thickness of about 50 ms TWT downslope, the thickness of Unit 9a decreases upslope to about 15 ms TWT. Shallow, near-seabed sediments from Unit 9a were sampled in cores taken more downslope, i.e. core ENAM93-21 and MD95-2009 (Rasmussen et al. 1996, 1998, 1999) (Fig. 1). An age of
PT
150,000 yrs BP has been found in the latter core for sediments present at a depth of about 25.5 m below seabed (Rasmussen et al., 1999).
RI
Figure 3 shows a section of a seismic line PSAT-217 at the northeastern extremity of the Fugloy
SC
Ridge (Fig. 1), illustrating the presence of sub-seabed mobilisation and formation of diapiric seabed features. The seismic record has been divided into six seismic units, named Unit A to F from below,
NU
separated by regional seismic horizons (Fig. 3; Kenyon et al. 2003). Unit A is interpreted to be of
MA
Paleocene age, Unit B and C of Eocene and Oligocene age, respectively, Unit D to be of Miocene age and Unit E and F to have been deposited during Plio-Pleistocene times (Kenyon et al, 2003 and
D
references herein). The diapirs are seen to originate from the underlying seismic Units D and C as
PT E
visualised by the chaotic, disturbed internal reflection patterns and disrupted top reflectors of these units. The diapirs have grown through seismic Units E and F to finally pierce the seabed. The two latter seismic units are part of the NSF system and can be correlated to the seismic units P4 and P1
CE
of Nygård et al. (2005). According to these authors, their units P4 and P1 consist of glacigenic
AC
debris flows and form the NSF sediment wedge at the northern outlet of the Faroe-Shetland Channel. Nygård et al. (2005) also noted diapiric features piercing through units P4 and P1 and could assign a MIS 6 age to unit P4. This indicates a (re-)activation of diapirism in our study area to have initiated during the MIS 6 (Saalian) glacial period.
4.2 Side-scan sonar and subbottom profiler records
8
ACCEPTED MANUSCRIPT
The side-scan sonar map shown in Fig. 4 illustrates the areal distribution of several mud diapirs piercing through surficial strata thus forming mounds at the seabed. The mosaic shows the piercing diapirs as isometric patches of strong backscatter (dark) and acoustic shadows (white) from steep flanks and extruded blocks. NSF sediments are characterised by relatively weak backscatter (light
PT
grey). Within the sedimentary succession, the sub-seabed diapiric structures are characterized by chaotic internal reflections (Figs. 3 and 5b). Diapirs with bathymetric expression above the sea floor
RI
may reach heights of up to 150 m above the surrounding seabed (Nielsen et al., 2002). An OKEAN
SC
long-range side-scan record acquired along seismic line PSAT-217 (Fig.3), having a 12 km swath range, demonstrate a more widespread occurrence of the seabed diapirs in the area (Kenyon et al.,
NU
2003). Thus, it seems that the mud diapirs are a relatively widespread seabed feature at the northern
MA
outlet of the Faroe-Shetland Channel. In our dataset, the areal distribution pattern of the diapirs does not provide clear evidence of a specific orientation in the area. Within this context it is,
D
however, noteworthy that the diapirs are all found within the southern rim of the Quaternary NSF
PT E
(Figs. 1 and 4).
4.3 Sediment coring
CE
During the TTR-12 cruise two gravity cores (AT359G, AT360G) and a dredge sample (AT361D)
AC
were obtained from selected diapir structures. The sediment sampling location is shown in Figures 4 and 5a. The precise coring position and water depth at the coring sites, together with results from AMS 14C dating of selected samples from these cores, are shown in Table 1. The 1.75 m long core AT359G was retrieved from a high-backscatter block observed on the MAK side-scan sonograph from the eastern flank of a c.50 m high diapir present in a diapir field more to the south (Fig. 5a). The 3.85 m long core AT360G was collected from the top of a c. 80 m high diapir structure in the northern diapir field. A simplified core lithology of these two cores is illustrated in Figure 6. With
9
ACCEPTED MANUSCRIPT
dredging AT361D sedimentary material was collected from the steep slope of the diapir in the northern part of the diapir field near coring site AT360G (Fig. 5a). A third core AT 358G (Fig. 5a) that was retrieved from the area outside, but nearby the diapirs, contained a 2 m thick sequence of glacigenic debris flows from the North Sea Fan.
PT
Core AT359G displayed fine-grained hemipelagic sediments in the upper part of the core. Lithified
RI
diatomic ooze was found at the base of the core (Fig. 7), which undoubtedly originates from the diapiric processes. Core AT360G revealed clear evidence of re-deposited sediment with slump
SC
structures, inclined and distorted boundaries, micro-faults, and contains clasts of more consolidated
NU
sediment. The presence of abundant diatomic ooze clasts embedded in the sediment below 140 cm in core AT360G and at the bottom of core AT359G (Figs. 6 and 7) may point to slumping caused
MA
by active diapirisme that extended at least into the last glaciation. At the dredging site AT361D only one sample was recovered, being a 0.20x0.15x0.15m large clast
D
of lithified light olive brown diatomic ooze of similar type as found in core AT360G. The results
PT E
from AMS 14C dating (Table 1) of selected samples from cores AT359G and AT360G demonstrate a late Weichselian age for the fine-grained, hemipelagic sediments found in these cores. Few
CE
samples analysed for their foraminiferal fauna (J.A. Rasmussen, unpubl. data) show a dominant planktonic fauna of Neogloboquadrina pachyderma (s). The dating of a sample from core AT360G
AC
(0.50-0.51 m core depth) yields an age of c. 25.640 14C yrs BP. In this core clasts of lithified diatomic ooze were found also at 141 cm core depth, which implies slumping of sediments on the flank of this diapir at some time prior to 25,000 years ago. A dinoflagellate analysis of the lithified diatomic ooze from this core and the dredge sample demonstrates a middle Miocene age for these samples (N.E. Poulsen, unpubl. data).
10
ACCEPTED MANUSCRIPT
5. Discussion 5.1 Iceberg turbate age and origin The origin of the chaotic internal reflections shown in the seismic line of Figure 2 has previously
PT
been attributed to deep-drafted iceberg grounding down to around 800 m present water depth (Nielsen and Van Weering, 1998). However, no estimate has been made on the age of the extreme
RI
deep-drafted iceberg grounding episode(s) that led to the formation of the turbate. In the following,
SC
we will focus on details of the seismic line shown in Figure 2, supplemented with relevant other information that may help to resolve the most likely age of formation of the iceberg turbate. As can
NU
be seen in Figure 2, the thickness of Unit 9a in the more downslope turbate area is about half the
MA
thickness of Unit 9 deeper down the slope where contourite deposition has prevailed. According to Rasmussen et al. (1999), the sediment age at about 11 m below seabed in core ENAM93-21 from the latter area (Fig. 1) is about 47.100 14C yrs BP. As described above and illustrated in Figure 2,
PT E
D
the top of the iceberg turbate underlies the regularly layered sediment cover of Unit 9a of which the thickness decreases upslope to about 15 ms TWT (Fig. 2). Using the information of the sediment type in Unit 9a to be a mixture of sand, silt and clay (Rasmussen et al. 1999), the sound velocity of
CE
Unit 9a should be around 1580 m/s (Hamilton and Bachman, 1981). This points to a thickness of
AC
the uppermost part of Unit 9a (Fig. 2) to be around 12 m. Considering an upslope decrease in sedimentation rate by a factor of 2½, which can be justified by comparing the age versus depths from cores ENAM93-21 and ENAM93-20 (Rasmussen et al. 2003; Fig. 1), the age level of 47.100 yrs BP in the more downslope turbate area would be reached at about 4.8 m below seabed. Assuming an average regular sedimentation rate of the uppermost Unit 9a (Fig. 2), this would imply a likely age of around 120,000 yrs BP for the top of the iceberg turbate, i.e. bottom of the uppermost Unit 9a layer, which corresponds to a time within the Eemian interglacial period (MIS
11
ACCEPTED MANUSCRIPT
5). Thus it seems most likely, that episodic iceberg reworking had been active until the beginning of MIS 5. A late deglacial drift of large icebergs towards the southeastern part of the Norwegian Sea may be supported by persisting low sea surface temperatures in the early Eemian recorded in sediment cores taken in this area (Rasmussen et al., 2003). The lower boundary of the iceberg
PT
turbate may correspond to the base of Unit 9, i.e. to a level of c. 100 ms TWT below the seabed in the contourite area (Fig. 2). Assuming a constant, but slightly increased average sound velocity of
RI
1600 m/s for Unit 9 due to compaction of the deeper lying sediments, it corresponds to a depth of
SC
80 m below seabed. Core MD95-2009 (Fig. 1), which sampled the contourite a little downslope of the outer limit of the turbate at 1027 m water depth (Fig. 1) gives an age of 150,000 yrs BP at 25.5
NU
m below the seabed (Rasmussen et al., 1999). Assuming a constant sedimentation rate, this yields
MA
for the base of Unit 9 an approximate age of around 470,000 yrs BP. Thus, formation of the iceberg turbate likely occurred within the period 120,000 to 470,000 yrs BP.
D
Glaciation records from Iceland report that the most extensive and severe glaciations all can be
PT E
dated within the period 140,000 to 500,000 yrs BP (Geirsdóttir et al., 2007; Van Vliet-Lanoë et al., 2010). The last (Weichselian) glaciation was notably less severe. The timing of the beginning of
CE
these severe glaciations corresponds to the findings of Ehlers and Gibbard (2007), who studied more generally the Quaternary history of Northern Hemisphere glaciations. With the help of
AC
evidence from other studies of extreme deep-draft iceberg grounding records from around the northern North Atlantic and Arctic Ocean we may further constrain the most likely age of the Faroe margin iceberg turbate formation. A recent study of ice-shelf grounding in the Arctic Ocean reports an age within MIS 6 (ca. 130,000-190,000 years ago) for extensive (> 1 km thick) ice shelf formation in the central Arctic (Jakobsson et al., 2016). This is supported by records of iceberg scouring at water depths in excess of 1 km in the Arctic near Fram Strait (Arndt et al., 2014). Also elsewhere in the Arctic, i.e. in Baffin Bay, evidence of extreme deep-draft (> 1 km) iceberg
12
ACCEPTED MANUSCRIPT
scouring associated with the MIS 6 glaciation has been found ( Kuijpers et al.,2007). Moreover, extreme deep-draft iceberg grounding likewise assigned to MIS 6 has also been observed on the southern flank of the Iceland-Faroe Ridge (Kuijpers and Werner, 2007). Studies of the Faroe Platform margin architecture and associated glacial features (Nielsen et al., 2007) demonstrate that
PT
a local ice stream is unlikely to have been the source of the deep-draft icebergs responsible for the turbate formation. With the various reports of more distant findings of extreme deep-draft iceberg
RI
drift associated with the MIS 6 (Saale) glaciation, we thus conclude that the latest episode of
SC
iceberg turbate formation on the upper slope of the North Faroe margin may be assigned to the MIS 6/5 transition. As shown by the North Atlantic distribution of ice-rafted debris associated with
NU
large-scale iceberg discharge from the North American Laurentide ice sheet (‘Heinrich Events’, e.g.
MA
Bond et al., 1992), and confirmed by the deep-draft iceberg scouring track on the Iceland-Faroe Ridge (Kuijpers and Werner, 2007), the dominating ocean current pattern during these events
D
resembled the modern regime. Another study of deep-draft iceberg plowmark trajectories (Newton
PT E
et al., 2016) shows that the flow of warmer, Atlantic-derived water along the Norwegian coast during the deglacial stage of MIS 12 still was northward, but that its speed was markedly reduced. The formation of the iceberg turbate on the northern Faroe margin may therefore be ascribed to drift
CE
of deep-draft icebergs originating from the Arctic, East Greenland and/or northern Iceland on their
AC
way to the southeastern Norwegian Sea, thus following an ocean current transport pathway comparable with the present circulation regime. This implies that a direct contribution from Scandinavia-derived icebergs is less likely, but the possibility of long-distance transport of such icebergs following the prevalent cyclonic ocean circulation pattern north of the Greenland-Scotland Ridge (Fig. 1) can not completely be excluded. 5.2 Diapir formation
13
ACCEPTED MANUSCRIPT
As described above, clasts of diatomic ooze of presumably middle Miocene age was found in core AT360G a little below the level where AMS 14C dating revealed an age of 25,000 yrs BP (Fig. 7). This suggests that diapir activity must have occurred at times during the last (Weichselian) glaciation. Johnson et al. (2005) made a regional study of the tectonic deformation of the area,
PT
which they found to have occurred since the Mid Miocene and according to these authors may have continued into recent times. Johnson et al. (2005) conclude that development of the diapiric
RI
structures found in the Faroe-Shetland Channel area may be the result of this large-scale, regional
SC
tectonic deformation. The location of the diapiric structures at the northern outlet of the FaroeShetland Channel occurs above a Cenozoic inversion dome (Long et al., 2003; Johnson et al.,
NU
2005), and marks the margin of the up to 1 km thick NSF. The NSF is, amongst others, made of
MA
several widespread glacigenic debris flow and mega-slide deposits (King et al., 1998; Nygård et al., 2005), of which the rapid emplacement and associated overpressure in the underlying hemipelagic
D
sediments must have led to subsurface sediment mobilisation. Similarly, faulting in fine-grained
PT E
abyssal plain sediments has been found as a result of sudden overloading due to rapid emplacement of thick (sandy) turbidites and debris flow deposits (Duin et al., 1985). The areal distribution of diapiric structures and mounds in our study area shows a clear relation with the position of the
CE
marginal zone of the NSF (Fig. 1) and is also located adjacent to where major turbidites originating
AC
from the northeastern Faroe margin are found (Kuijpers et al., 2001). Since the last interglacial at least 4 large-scale mass flow events occurred in this sector of the Faroe margin, with one major event dated at time of the LGM (Kuijpers et al., 2001). Moreover, timing of major depositional events on the NSF can be correlated with the same period, while ending close to 15,000 yrs BP as the Norwegian Channel had become free of ice (King et al., 1998; Haflidason et al., 1998). The AMS 14C information from core AT360G confirms diapiric activity coinciding with the last glaciation.
14
ACCEPTED MANUSCRIPT
Repeated deposition of heavy glacigenic mass flow deposits upon non-consolidated, relatively lowdensity, fine-grained hemipelagic ooze sediments in response to NSF progradation (Stoker et al., 2005) apparently caused (re)activation of sub-seabed sediment mobilisation and diapiric processes. The sediment mobilization process eventually leading to the diapir formation may have had its
PT
origin in the regional tectonic deformation described by Johnson et al. (2005). However, the seismic and side-scan sonar records illustrated in Figures 3 to 5, together with the coring results from the
RI
area, support that an intensification of more explicit diapiric activity occurred as the NSF deposits
SC
reached the deeper area of the northeastern Faroe margin in the MIS 6 glacial period.
MA
NU
5. Conclusions
Based on high-resolution seismic data from the upper slope of the northern Faroe margin and
D
previously published sediment core information from that area, the age of extensive deep-draft
PT E
iceberg grounding episodes at (present) water depths down to 800 m is determined at between 470,000 and 120,000 yrs BP. With further information from other sources, we conclude that the
CE
most likely age of the latest episode was at the MIS6/5 transition. The seismic and acoustic data from the diapir study area suggest that repeated emplacement of
AC
heavy glacigenic North Sea Fan mass flow deposits upon relatively low-density, fine-grained hemipelagic ooze sediments caused (re)activation of sub-seabed sediment mobilisation and diapiric processes. On the north Faroe Margin the diapirism intensified in the MIS 6 glacial period as the North Sea Fan deposits reached the area. One of the more recent stages of activation can be dated to have occurred at a time during the last (Weichselian) glaciation. An initial mobilisation of the ooze sediments may, however, be a result of regional tectonic deformation since Mid Miocene times,
15
ACCEPTED MANUSCRIPT
with diapiric processes being intensified in the Pliocene-Pleistocene as Northern Hemisphere glaciations became more severe.
Acknowledgements
PT
The study leading to this paper forms part of the GLANAM (GLAciated North Atlantic Margins) Initial Training Network funded by the People Programme (Marie Curie Actions) of the European
RI
Union’s Seventh Framework Programme FP7/2007-2013/ under REA grant agreement n° 317217.
SC
The data of this study was collected during 2 cruises of R/V Pelagia and one cruise (TTR-12) of
NU
R/V Prof. Logachev. The latter cruise was part of the international, UNESCO supported ‘TraningTrough-Research’ (TTR) programme. Early research has financially been supported by GEUS, the
MA
EU 7th Framework Programme (ENAM project) and the Faroese GEM/FOIB offshore consortium, respectively. We greatly acknowledge the engagement of Master and crew of the two research
D
vessels as well as the efforts of the responsible principal scientists (Tjeerd C.E. van Weering, Royal
PT E
NIOZ; late Michael Ivanov, Moscow State University) during the work at sea. In addition, we thank technicians and many students (e.g. Moscow State University) having been involved in the work at
CE
sea and in the following laboratory investigations. Finally, we acknowledge the two reviewers
AC
whose comments were very useful for the improvement of our manuscript.
16
ACCEPTED MANUSCRIPT
References
Arndt, J.E., Niessen, F., Jokat, W., Dorschel., B., 2014. Deep water paleo-iceberg scouring on top
PT
of Hovgaard Ridge – Arctic Ocean. Geophys. Res. Lett. 41, 2014GL060267.
Bond, G., Heinrich, H., Labeyrie, L., McManus, J., Andrews, J., Huon, S., Jantschik, R., Clasen, S.,
RI
Simet, C., Tedesco, K., Klas, M., Bonani, G., Ivy, S., 1992. Evidence of massive discharges of
SC
icebergs into the North Atlantic Ocean during the last glacial period. Nature 360, 245–249.
NU
Duin, E.J.Th., Mesdag, C.S., Kok, P.T.J., 1985. Faulting in Madeira Abyssal Plain sediments. Mar.
MA
Geol. 56 (1-4), 299–308.
D
Ehlers, J., Gibbard, P.L., 2007. The extent and chronology of Cenozoic Global Glaciation. Quat.
PT E
Intern. 164-165, 6–20.
Geirsdóttir, A., Miller, H., Andrews, J., 2007. Glaciation, erosion, and landscape evolution of
AC
CE
Iceland. J. Geodyn. 43 (1), 170–186.
Haflidason, H., King, E. L., Brett, C. P., Stevenson, A. G., Wallis, D. G., Campbell, N., Sejrup, H. P. & Waage, B. 1996. Marine geological/geophysical cruise report on the North Sea margin. Upper North Sea Fan, Miller Slide and Faeroe-Shetland Channel. R/V Haakon Mosby cruise No. HM11096. University of Bergen.
Haflidason, H., King, E.L., Sejrup, H.P., 1998. Late Weichselian and Holocene sediment fluxes of
17
ACCEPTED MANUSCRIPT
the northern North Sea Margin. Mar. Geol. 152 (1-3), 189–215.
Hansen, B., Østerhus, S., 2000. North Atlantic – Norwegian Sea exchanges. Prog. Oceanogr. 45,
PT
109–208.
Hansen, B., Østerhus, S., Dooley, H.D., Gould, W.J., Rickards, L.J., 1998. North Atlantic –
SC
RI
Norwegian Sea exchanges: The ICES NANSEN Project. ICES Coop.Res. Rep. 225.
Hubbard, A., Sugden, D., Dugmore, A., Norddahl, H., Pétursson, H.G., 2006. A modelling insight
MA
NU
into the Icelandic Last Glacial Maximum ice sheet. Quat. Sci. Rev. 25, 22832296.
Humlum, O., Christiansen, H.H., Svensson, H., Mortensen, L. E., 1996. Moraine Systems in the
PT E
D
Faroe Islands: Glaciological and Climatological Implications. Geogr. Tidskr. 96, 21–31.
Jakobsson, M., Nilsson, J., Anderson, L., Backman, J., Björk, G., Cronin, T.M., Kirchner, N., Koshurnikov, A., Mayer, L., Noormets, N., O’Regan, M., Stranne, C., Ananiev, R., Macho, N.B.,
CE
Cherniykh, D., Coxall, H., Eriksson, B., Flodén, T., Gemery, L., Gustaffson, Ö., Jerram, K.,
AC
Johansson, C., Khortov, A., Mohammad, R., Semiletov, I., 2016. Evidence for an ice shelf covering the central Arctic Ocean during the penultimate glaciations. Nature Comm. DOI: 10.1038/ncomms10365.
Johnson, H ., Ritchie, J .D., Hitchen, K ., McInroy, D .B., Kimbell, G .S., 2005.Aspects of the Cenozoic deformational history of the Northeast Faroe–Shetland Basin,Wyville–Thomson Ridge and Hatton Bank Areas. In: Doré, A.G.,Vining, B.A. (Eds.), Petroleum Geology: North-West
18
ACCEPTED MANUSCRIPT
Europea and Global Perspectives, Proceedings 6th Petroleum Geology Conference, Petroleum Geology Conferences Ltd., Geological Society, London, 993–1007.
Jørgensen, G., Rasmussen, J., 1986. Glacial Striae, roches moutonnées and ice movements in the
PT
Faroe Islands. Geological Survey of Denmark, DGU Series C, 7, 114 pp.
RI
King, E.L., Haflidason, H., Sejrup, H.P., Løvlie, R., 1998. Glacigenic debris flows on the North Sea
SC
Trough Mouth Fan during icestream maxima. Mar. Geol. 152 (1-3), 217–246.
NU
Kuijpers, A., Nielsen, T., Akhmetzhanov, A., De Haas, H., Kenyon, N.H., Van Weering, T.C.E.,
MA
2001. Late Quaternary slope instability at the Faeroe margin: mass flow features and timing of
D
events. Geo-Mar. Lett. 20, 149–159.
PT E
Kuijpers, A., Werner, F., 2007. Extremely deep-draft iceberg scouring in the glacial North Atlantic. Geo-Mar. Lett. 27, 383–389.
CE
Kuijpers, A., Dalhoff, F., Brandt, M.P., Hümbs, P., Schott, T., Zotova, A., 2007. Giant iceberg plow
AC
marks at more than 1 kilometre water depth offshore West Greenland. Mar. Geol. 246, 60–64.
Long, D., Hoult, R., Haflidason, H., Stevenson, A.G., Ritchie, J.D., Johnson, H., De Haas, H., Van Weering, T.C.E., Mienert, J., 2003. Mud Mound/?Diapiric Features in the Faroe - Shetland Channel. Geophys. Res. Abstr. 5, 11201, European Geophysical Society Assembly, France, 2003.
19
ACCEPTED MANUSCRIPT
Newton, A.M.W., Huuse, M., Brocklehurst, S.H., 2016. Buried iceberg scours reveal reduced North Atlantic Current during the Stage 12 deglacial. Nat. Comm., DOI: 10.1038/ncomms10927.
Nielsen, T., Weering, T.C.E. van, 1998. Seismic stratigraphy and sedimentary processes at the
PT
Norwegian Sea margin northeast of the Faroe Islands. Mar. Geol. 152 (1-3), 141–157.
RI
Nielsen, T., Kuijpers, A., Ivanov, M. And Shipboard Scientific Party (2002). Mapping and
SC
sampling diapirs and mass flows on the Eastern Faroe Margin. Geological Survey of Denmark and
NU
Greenland Report 2002/75, 23 pp.
MA
Nielsen, T., Kuijpers, A., 2004. Geohazard studies offshore the Faroe Islands: slope instability, bottom currents ans sub-seabed sediment mobilization. Geological Survey of Denmark and
PT E
D
Greenland Bulletin 4, 57-60.
Nielsen, T., Rasmussen, T.L., ceramicola, S., Kuijpers, A., 2007. Quaternary sedimentation, margin
AC
1016–1036.
CE
architecture and ocean circulation variability around the Faroe Islands. Quat. Sci. Rev., 26 (7-8),
Nygård, A., Sejrup, H.P., Haflidason, H. and Bryn, P. (2005). The glacial North Sea Fan, southern Norwegian Margin: architecture and evolution from the upper continental slope to the deep-sea basin. Marine and Petroleum Geology, 22, 71–84
Rasmussen, T.L., Balbon, E., Thomsen, E., Labeyrie, L., van Weering, C.E., 1999. Climate records and changes in deep outflow from the Norwegian Sea ~150-55 ka. Terra Nova, 11, 61-66.
20
ACCEPTED MANUSCRIPT
Rasmussen, T.L., Thomsen, E., Kuijpers, A., Wastegaard, S., 2003. Late warming and early cooling of the sea surface in the Nordic seas during MIS 5e (Eemian Interglacial). Quat. Sci.Rev. 22, 809–
PT
821
Rasmussen, T.L., Van Weering, T.C.E., Labeyrie, L., 1996. High resolution stratigraphy of the
RI
Faeroe-Shetland Channel and its relation to North Atlantic paleoceanography: the last 87 kyr.
SC
Mar. Geol. 131, 75–88.
NU
Ritchie, J.D., Johnson, H. & Kimbell, G.S. 2003. The nature and age of Cenozoic contractional
MA
deformation within the NE Faroe-Shetland Basin. Marine and Petroleum Geology, 20, 399–409.
D
Rasmussen, T.L., Wastegaard, S., Kuijpers, A., van Weering, T.C.E., Heinemeier, J., Thomsen, E.,
PT E
2003. Stratigraphy and distribution of tephra layers in marine sediment cores from the Faeroe Islands, North Atlantic. Mar. Geol. 199, 263-277.
CE
Stoker, M.S., Praeg, D, Shannon, P.M., Hjelstuen, B.O., Laberg, J.S., Nielsen, T., 2005: Neogene
AC
stratigraphy and sedimentation on the NW European Atlantic margin. Marine and Petroleum Geology, 22, 977-1005.
van Weering, T.C.E., Nielsen, T., Kenyon, N.H., Akentieva, K., Kuijpers, A., 1998. Large Submarine slides at the NE Faeroe continental margin. In: Stoker, M.S., Evans, D., Cramp, A. (eds.): Geological Processes on Continental Margins: Sedimentation, Mass-Wasting and Stability. Geological Society, London, Special Publications, No. 129, 5-17.
21
ACCEPTED MANUSCRIPT
Vliet-Lanoë, B. van, Gudmundsson, A., guillou, H., Guégan, S., Loon, A.J. van, De Vleeschouwer,
AC
CE
PT E
D
MA
NU
SC
RI
PT
F., 2010. Glacial Terminations II and I as recorded in NE Iceland. Geologos, 16 (4), 201–222.
22
ACCEPTED MANUSCRIPT
Figure and Table Captions Fig. 1 – Regional bathymetric map of the North Faroe margin (compiled by W. Weng, GEUS) with location of seismic and side-scan sonar tracks and sampling positions. For regional setting, see inset map. Also shown is the outline of the North Sea Fan (greyish; after Nygård et al. 2005), the iceberg
PT
turbate (whitish; after van Weering et al., 1998) and the major slide scar of the North Faroe Slide
RI
Complex (NFSC; after van Weering et al. 1998). Reference to following figures is marked in red.
SC
Fig. 2 – Part of high-resolution multi-channel seismic line PE91-06 on the upper slope of the North Faroe margin. Shelfwards, the chaotic reflection pattern of the iceberg turbate (yellow) is seen,
NU
while downslope the parallel seismic reflections of the contourite are seen (modified from Nielsen
MA
and van Weering, 1998). In the transition zone, the iceberg turbate is overlain by the upper part of
D
the Unit 9 contourite deposits, here termed Unit 9a. For location, see Figure 1.
PT E
Fig. 3 - High-resolution single-channel seismic line PSAT-217 across the Fugloy Ridge showing interpreted seismic units (A-F), seismic evidence of sub-seabed sediment mobilisation, and seabed diapir features (see text for details). Seismic units E and F represent the North Sea Fan. For
AC
CE
location, see Figure 1.
Fig. 4 – Mosaic image of the deep-towed MAK side scan sonar lines crossing the mud diapirs. Insert map shows location of the two lines composing the mosaic, and their location with respect to seismic line PSAT-217 (blue) shown in Fig. 3. Location of sampling sites referred to in text and shown in Fig. 5 is also indicated. The diapirs are seen as strong backscatter (dark) and acoustic shadows (white). North Sea Fan sediments are characterised by relatively weak backscatter (light
23
ACCEPTED MANUSCRIPT
grey). The white line represents the track of the deep-towed side-scan vehicle. Side-scan sonar range is 2 km across each line. For mosaic site location, see Figure 1.
Fig. 5 Deep-towed side scan sonar line MAK-AT46 crossing the diapirs (for location of line and
PT
sampling sites, see insert map in Fig. 4). a) Sonograph with location of the coring and dredge sites TTR12-AT358G, AT359G, AT360G (red), and AT361D (yellow) showing piercing diapirs as
RI
isometric patches with high and medium backscatter. b) Corresponding subbottom profile showing
SC
cross-sections of the diapirs and glacigenic debris flow lobes of the North Sea Fan (NSF). m.b.s.s =
NU
meter below sea surface. For regional setting of line, see Figure 1.
MA
Fig. 6 – Simplified lithological columns and legend for cores AT359G and AT360G (after Kenyon et al. 2003). For core locations, see Figs 4 and 5. The figure also shows results from radiocarbon
PT E
D
dating (AMS 14C yrs BP) and magnetic susceptibility (counts) measurements.
Fig. 7 – Photo of the bottom 31 cm of core AT359G (see Fig. 6) showing Miocene diatomic ooze
CE
(for details, see text). Core diameter is 10 cm.
AC
Table 1 – Sediment sampling positions and AMS 14C results.
24
ACCEPTED MANUSCRIPT
Utc Lab. No.
AMS 14C age
0.17-0.18 m
12186
16,820±90
0.41-0.43 m
12187
38,400±700
1548 m
0.50-0.51 m
12188
25,640±180
1640 m
_
PT
Table 1 Site No.
Position
Water depth
14
AT359G
62o38,712N
1558 m
_
1o 17,145W AT360G
62o41,409N
C sample
1o15,835W
_
RI
62o36,002N
CE
PT E
D
MA
NU
SC
1o19,016W
AC
AT359D
25
ACCEPTED MANUSCRIPT
Highlights Large icebergs from Greenland and Arctic formed turbate on upper north Faroe margin
The iceberg turbate formation occurred between 470.000 and 120.000 yrs BP
The latest extensive episode of turbate formation occurred at the MIS6/5 transition
Diapirism in deep waters off NE Faroes Island likely caused by density inversion
(Re-)activation of diapirism in MIS6 due to North Sea Fan glacigenic deposits
The most recent stage of diapir activity dates to the period of the LGM
AC
CE
PT E
D
MA
NU
SC
RI
PT
26
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7