Submarine landslides and tsunami threat to Scotland

Abstract

There is strong evidence that the Storegga Slide located offshore mid-Norway caused a tsunami wave that struck the northern and eastern coasts of Scotland about 7200 years ago. Its impact physically was extensive, but socio-economically we presume it was minor. Today its consequences would be very different. Other landslides, ranging in volume from 0.2 km 3 to more than 800 km 3 have been identified on the continental margin around Scotland and need to be evaluated as to the risk the repetition that such events pose. Work is underway to map and date these events, and assess their potential triggers. The risk assessment includes evaluation of offshore seismicity and the geotechnical parameters of slope sediments. Slide frequency during the Quaternary and the environment of failure are also important factors. These marine studies are matched by studies onshore for evidence of paleotsunamis.

Full text

ITS 2001 Proceedings, Session 1, Number 1-12 355
Submarine landslides and tsunami threat to Scotland
David Long and Richard Holmes
British Geological Survey, Edinburgh, Scotland, United Kingdom1
Abstract. There is strong evidence that the Storegga Slide located offshore mid-Norway caused
a tsunami wave that struck the northern and eastern coasts of Scotland about 7200 years ago.
Its impact physically was extensive, but socio-economically we presume it was minor. Today its
consequences would be very different. Other landslides, ranging in volume from 0.2 km3to more than
800 km3have been identified on the continental margin around Scotland and need to be evaluated
as to the risk the repetition that such events pose. Work is underway to map and date these events,
and assess their potential triggers. The risk assessment includes evaluation of offshore seismicity
and the geotechnical parameters of slope sediments. Slide frequency during the Quaternary and
the environment of failure are also important factors. These marine studies are matched by studies
onshore for evidence of paleotsunamis.
1. Regional Setting
Scotland sits on the northwest European passive margin, an area not usu-
ally considered affected by the more active geohazards found in many other
parts of the world. Following the opening of the North Atlantic in late Meso-
zoic/early Paleogene time and the associated voluminous volcanic activity
associated with the Iceland Plume, the traditional image of NW Europe is of
an inactive area. As the Tertiary volcanics cooled and subsided the uplifted
areas were subjected to rapid erosion, forming extensive basin deposits, some
exploited for their hydrocarbons in the North Sea and nearby. Denudation
rates fell until the Neogene when centers of uplift developed in Norway, the
Faroes, Scotland, and Ireland in response to crustal stress caused by the
Alpine Orogeny. The denudation rates increased further during the Quater-
nary when extensive ice sheets periodically developed and retreated on these
uplifted areas. The eroded sediments were deposited at or just beyond the
shelf break, in places advancing it 20 km in less than half a million years.
The distribution of the eroded sediment was focused on selected areas
where up to 1 km of Plio-Pleistocene sediments have been mapped (Fig. 1).
The distribution of the depocenters reflects the sediment transport pathways
across the continental shelf, particularly where sedimentary basins provide
more easily erodible surfaces compared with that of early Paleozoic and older
basement. The most southerly depocenter comprising the Barra and Donegal
Fans, subdivided by the Hebrides Terrace Seamount, is up to 900 m thick
between 56◦Nand57
◦N extending from the shelf break (∼200 m) to more
than 2000 m water depth (Fig. 1). The next depocenter, the Sula Sgeir Fan,
occurs at the northeast corner of the Rockall Trough, where up to 600 m of
sediments have been mapped. Like the Barra Fan these sediments comprise
a large sediment wedge prograding from the shelf to the floor of the Rockall
Trough. In contrast, in the between-fans area only a thin (<100 m thick)
1British Geological Survey, West Mains Road, Edinburgh EH9 3LA, Scotland, United
Kingdom (D.long@bgs.ac.uk, r.holmes@bgs.ac.uk)

356 D. Long and R. Holmes
12W
11W
10W
9W
8W
7W
6W
5W
4W
3W
2W
1W
0E
1E 2E 3E 4E 5E
6E
7E
8E
54N
55N
56N
57N
58N
59N
60N
61N
62N
63N
64N
Thickness of Plio-
Pleistocene in metres
50
100
200
400
600
800
>800
Barra
Fan
Sula Sgeir
Fan
Ireland
Scotland
Norway
Shetland
Faroes
Faroe-ShetlandChannel
North Sea Fan
Figure 1: Map showing Plio-Pleistocene thickness along the Scottish margin; crosses mark centers of
Pliocene uplift.

ITS 2001 Proceedings, Session 1, Number 1-12 357
NW SE
(Not to scale)
(Not to scale)
FAROE-SHETLAND
CHANNEL
FAROE-SHETLAND
CHANNEL
Mostly
undivided
Mostly
undivided
WEST SHETLAND SLOPE
WEST SHETLAND SLOPE
WEST SHETLAND SHELF
WEST SHETLAND SHELF
Glacial Pleistocene
Pre-glacial Pleistocene
Pliocene
Miocene
Figure 2: Schematic cross-section showing the sediment wedges build out west of Shetland.
Plio-Pleistocene sequence has accumulated. Seismic profiles suggest that in
the fans a wide range of along-slope and down-slope sedimentary processes
have been active, however in the inter-fan areas the sedimentary processes
appear to be restricted to contourite sheets.
In the Faroe–Shetland Channel the sediment thicknesses are less and not
as focused, reflecting perhaps the greater distance from mainland Scotland,
and lower uplift and more limited glaciation over Shetland. Sediments locally
exceed 200 m thick, extending the shelf break 20 km within the glacial Qua-
ternary (Fig. 2). However, to the north and northeast of Shetland the North
Sea Fan has the greatest thickness of Plio–Pleistocene sediments originating
from Scandinavia with minor amounts from the UK. Much of the Scandina-
vian ice sheet flowed into the Norwegian Trench where it turned northwards
to shed its load onto the North Sea Fan. This fan records the largest volume
of geologically recent uncemented sediments on the European margin with
25,000 km3of Plio–Pleistocene sediments, locally exceeding 1200 m thick,
making it comparable to the fan of a major river system.
2. Evidence of Sediment Failure
Much of the UK margin has been systematically surveyed as part of the
British Geological Survey’s regional mapping program. There have also been
selected, more detailed, studies for various academic projects and numerous
commercially funded site surveys associated with hydrocarbon exploration.
The Barra Fan shows evidence of numerous debris flows giving a chaotic
acoustic appearance between regionally extensive reflectors. The latter are
cut by erosive events which, when mapped out, depict several large slides.
These are known collectively as the Peach Slides and together displace
1830 km3of sediment (Holmes et al., 1998). Definitive dates for these
events are not available but event 3 is less than 17,000 years BP and event
4 intersects iceberg ploughmarked seabed, suggesting a late- to post-glacial
age. High resolution swath bathymetry and sidescan sonar data also show
smaller superficial sediment movement with very fresh appearances, partic-
ularly within the Irish sector on the Donegal Fan (Fig. 3). These Holocene
failed sediments are predominantly at the shelf margin.
Farther north, the Sula Sgeir Fan shows downslope sediment movement

358 D. Long and R. Holmes
1980
3.3ML 1986
3.1ML
12
2
3
4
5
10°W
57°N
56°N
9°W
11°W
200
200
1000
1000
1500
1500
2000
2000
1500
1500
1000
1000
1000
1000
1500
1500
2000
2000
2000
2000
1500
1500
1000
1000
200
200
N
20km
Donegal Fan
Donegal Fan
Barra Fan
Barra Fan
Peach Slide
Peach Slide
Hebrides Terrace
Seamount
Hebrides Terrace
Seamount
Figure 3: Map of the Barra and Donegal fans. Light grey—area of Plio-Pleistocene
debris deposits, mid-grey—Peach Slide events, dark grey—recent events, stars—
locations of seismic events with magnitudes.

ITS 2001 Proceedings, Session 1, Number 1-12 359
2°30’W 2°26’W 2°22’W
61°20’N
61°18’N
61°16’N
2km
2km
N
Figure 4: Seabed image showing downslope debris flows and base of slope fans.
in several forms. GLORIA surveys identified three bottle-neck slides, 1–
2 km wide, 10 km long (Kenyon, 1987). These were originally interpreted as
post-glacial as their headwalls appeared to have cut iceberg ploughmarks.
However, more recent surveys show they are older and are just the most re-
cent of numerous events extending back to when shelf-wide glaciation began.
Along the west Shetland margin there is abundant geomorphological ev-
idence of numerous debris flows with base of slope fans. However, they all
appear to be glacigenic and are located downslope of where ice sheets ex-
tended beyond the shelf break (Fig. 4). Toward the northern end of the
Faroe–Shetland Channel, seabed surveys show clear evidence of a recent
landslide (Fig. 5). This feature, the Afen Slide, 3 km wide, 13 km long,
has been tentatively dated to mid-Holocene. Detailed studies show that it
is a multi-phase feature with possible retrogressive failure upslope. There
has also been some sidewall failure. The total volume of displaced sediment

360 D. Long and R. Holmes
10km
10km
N
5°W 4°48’W
4°48’W
4°36’W
4°36’W
4°24’W
4°24’W
4°12’W
4°12’W
60°10’N
60°10’N
60°30’N
60°30’N
60°20’N
60°20’N
5°12’W
5°12’W
Figure 5: Seabed image of the Afen Slide.
involved in this feature is about 0.4 km3. Recent work has identified an-
other slide of virtually the same dimensions buried about 50 m below the
present seabed, seismo-stratigraphically several 100,000 years old. To the
north, another buried slide, the Miller Slide, has a headwall up to 100 m
high and a debris flow extending more than 100 km out into the Faroe–
Shetland Channel (Fig. 6). This may have displaced more than 200 km3of
slope sediments. Seismo-stratigraphic correlation suggests an O18 stage 9 or
11 age to this feature. This slide is located close to the edge of the North Sea
Fan, within which there is abundant evidence for large buried events (King
et al., 1996; Evans et al., 1996). They include syndepositional debris flows
associated with glacial processes and landslides that have transported large
blocks of sediment (Fig. 7). Most of these features are within the Norwegian
sector of the northern North Sea and the modern seabed-failure analogue is
the 7200 year BP Storegga Slide located on the northern flank of the North
Sea Fan.
3. Triggers
The rapid sedimentation in selected loci along the UK margin has cre-
ated thick sequences of under-consolidated sediments. Excess pore-pressures
within the sediment pile are presumed to occur due to this rapid loading.

ITS 2001 Proceedings, Session 1, Number 1-12 361
Miller Slide deposit
1:15 Vertical Exaggeration
2km
500
600
700
800
900
1000
110 0
1200
Depth below sea level (in metres)
Base Plio-pleistocene
SE16/6016/6516/7016/75NW
Slope angle
0.5°
1.0°
2.0°
3.0°
4.0°
5.0°
1400
1500
1600
1700
1800
1900
2000
Depth below sea level (in metres)
Tertiary mass flow unit
Top Eocene
Merges
with other
debris deposits
1500
1000
500
200
1500
2°W 1°W
62°30’N
62°N
61°30’N
Figure 6: The Miller Slide northwest of Shetland—interpreted seismic section and
location map
SEA BED
SEA BED
NNW
NNW
SSE
SSE
DEBRIS-FLOW LENSES
DEBRI S-FL OW LEN SES
mid to late Pleistocene
mid to la te Ple isto cene
1600 ms
1600 ms
{
FLOW OVER VIGRA BLOCK
FLOW OVER VIGRA B LOCK
MØRE SLIDE SEDIMENTS
MØRE SLID E SEDI MENT S
~100 m
~100 m
VIGRA SLIDE BLOCK
VIGRA S LIDE BLO CK
VIGRA GLIDEPLANE
VIGRA G LIDEPLAN E
0 5 10
10
kilometres
kilom etre s
200 ms
200 ms
(~180 m in sediment)
(~180 m in s edim ent)
TAMPEN SLIDE GLIDEPLANE
TAMPEN SL IDE GLI DEPL ANE
TOP OF TAMPEN SLIDE
TOP OF TAMPE N SLID E
disturbed sediments
distu rbed s edim ents
MØRE SLIDE GLIDEPLANE
MØRE SLID E GLID EPLA NE
VIGRA SLIDE SCARP
VIGRA S LIDE SCA RP
VIGRA SLIDE GLIDEPLANE
VIGRA S LIDE GLI DEPLANE
SEDIMENTS
SEDIM ENTS
DISTURBED
DISTU RBED
MØRE SLIDE
MØRE SLID E
HEMIPELAGIC
HEMIP ELAG IC
Figure 7: Seismic section from the North Sea Fan showing displaced blocks
(adapted from Evans et al., 1996).

362 D. Long and R. Holmes
Gas may also contribute to the excess pore-pressure due to the breakdown
of in-situ organic matter and leakage from underlying hydrocarbon reser-
voirs. Acoustic gas blanking has been noted with some of the slide features.
However, many of the sites occur within the methane hydrate stability zone
and so, except in areas of high gas flux, free gas may not occur. Other
than some evidence within the Storegga complex, bottom simulating re-
flectors and other acoustic evidence for hydrates are absent. As the NW
European margin has been subjected to large eustatic and isostatic tectonic
movements during the Quaternary, significant pressure changes (and thermal
fluctuations) may have sublimated much of any hydrate present.
Seismicity is normally low along a passive margin. However, the northern
end of the North Sea is one of the more active areas of NW Europe. There
have been 90 events of magnitude >3M
Lin the last 30 years out of 1500
recorded by stations in the UK and Norway (2 events >magnitude 5 ML).
West of the UK, monitoring has been more limited, but two events of mag-
nitude >3M
Lwere detected in the 1980s close to the Peach Slide (Fig. 3).
Activity in the Faroe–Shetland Channel has been monitored over the last 5
years without detection (current detection threshold 2 to 2.5 ML). However,
the location of the Afen Slide, above a significant tectonic lineament, the
Victory Transfer Zone, mimics the co-location of the Storegga Slide above
the Jan Mayen Fracture Zone and the Trænadjupet Slide above the Bivrost
Lineament offshore Norway (Laberg and Vorren, 2000). It should be noted
that modern seismic activity may be lower than that in the early Holocene
when the postglacial crustal rebound rate was greater. Neotectonism is ev-
ident at glacial centers in Scotland (Ringrose, 1989) and Northern Ireland
(Knight, 2000) with surface displacement since deglaciation.
4. Threats to Scotland
All these slides on the continental margin are located more than 70 km from
the coastline, therefore only the largest sea perturbations are likely to impact
the coast. The western and northern coastlines of Scotland are sparsely pop-
ulated, however a few key economic sites are potentially vulnerable. Perhaps
of greater impact, although originating from much smaller events, would be
submarine landslides in the sea lochs of western Scotland. The steeper sides,
greater late- to post-glacial sedimentation rates, and elevated seismicity due
to post-glacial crustal rebound make these areas worthy of further study.
The Afen Slide of mid-Holocene age is located above a similar feature
indicating repetition. The most recent Peach Slide is the latest of two post-
dating 17 ka. The North Sea Fan has had repeated failure. Together this
indicates that the threat of new landslides is ever present. The larger events
such as the Peach and Miller slides, if displaced singularly, might have caused
a tsunami. They are as large as some other slides that have been associated
with tsunamis over distances as great as that between the slide and the
present day coastlines. However, due to their suspected age it is extremely
unlikely that any geological evidence exists to confirm this. Along the eastern

ITS 2001 Proceedings, Session 1, Number 1-12 363
Figure 8: Map of tsunami deposits attributed to the Storegga Slide. Solid dots—sites dated to about 7200
yBP, open dots—sites undated.

364 D. Long and R. Holmes
8
7
6
5
4
3
2
1
0
-1
-2
-3
0100 200 300 400 500 540
Metres
WE
Colluvium
Estuarine clays
Estuarine clays
Tsunami sand deposit
Peat
Late glacial
deposits
Metres
Figure 9: Profile at Creich, east coast of Scotland, illustrating the litho-
transgressive nature of the Storegga tsunami deposit. For location see Fig. 8.
Figure 10: Photograph of tsunami sand layer (behind shaft of spade) within peat
deposits at Maryton, east coast of Scotland. For location see Fig. 8.
and northern coasts of Scotland, though, there is evidence for a tsunami
associated with the Storegga Slide of 7200 yBP (Dawson et al., 1987).
The landslide’s impact on Scotland was to cause a tsunami that struck
the north and eastern coasts extending as far south as Lindesfarne in north-
ern England (Fig. 8). Based on the sedimentological evidence, the waves
would locally have extended several hundred meters inland of the former
coastline with a run-up of 1–2 m in open areas and much greater in enclosed
bays or lochs (Long et al., 1989). These figures are based on sediments laid
down (Fig. 9) and subsequently preserved, therefore representing minimum
run-up values. The tsunami sediments (Fig. 10) typically comprise marine
sands but also contain debris from the coastal marshes, etc. Detailed ex-
amination of this debris indicates that the event happened in the autumn
(Dawson and Smith, 2000), matching similar evidence in Norway (Bonde-

ITS 2001 Proceedings, Session 1, Number 1-12 365
vik, 1997) where recent high precision dating gives an age of 7262 ±47 yBP
(Bondevik, personal communication, 2001). We have to presume the human
impact was small due to the low population levels 7200 years ago, yet the de-
posit has been found at sites of early human habitation. However, we should
presume that if it occurred today the consequences would be economically
catastrophic. The frequency of tsunamis can be considered extremely low
but not non-existent and needs to be considered in long-term planning for
Scotland.
5. Current Work
Continued mapping for landslides offshore and tsunami deposits onshore
continues under a range of oil company, European, and national funded
programs.
Acknowledgments. This paper forms part of a work under the COSTA-Europe
program funded by the EU. The Western Frontiers Association has funded seabed
imagery and landslide studies. The authors publish with permission of the director,
British Geological Survey.
6. References
Bondevik, S., J.I. Svendsen, G. Johnsen, J. Mangerud, and P.E. Kaland (1997):
The Storegga tsunami along the Norwegian coast, its age, and runup. Boreas,
26, 29–53.
Dawson, A.G., D. Long, and D.E. Smith (1988): The Storegga slide; evidence from
eastern Scotland for a possible tsunami. Mar. Geol., 82, 271–276.
Dawson, S. and D.E. Smith (2000): The sedimentology of Middle Holocene tsunami
facies in northern Sutherland, Scotland, UK. Mar. Geol., 170, 69–79.
Evans, D., E.L. King, N.H. Kenyon, C. Brett, and D. Wallis (1996): Evidence
for long term instability in the Storegga Slide region off western Norway. Mar.
Geol., 130, 281–292.
Holmes, R., D. Long, and L.R. Dodd (1998): Large-scale debrites and submarine
landslides on the Barra Fan, west of Britain. In Geological Processes on Con-
tinental Margins: Sedimentation, mass-wasting and stability,editedbyM.S.
Stoker, D. Evans, and A. Cramp, Geological Society, London, Special Publica-
tions, 129, 67–79.
Kenyon, N.H. (1987): Mass-wasting features on the continental slope of northwest
Europe. Mar. Geol., 74, 57–77.
King, E.L., H.P. Sejrup, H. Haflidason, A. Elverhøi, and I. Aarseth, (1996): Qua-
ternary seismic stratigraphy of the North Sea Fan: glacially fed gravity flow
aprons, hemipelagic sediments, and large submarine slides. Mar. Geol., 130,
293–315.
Knight, J. (1999): Geological evidence for neotectonic activity during deglaciation
of the southern Sperrin Mountains, Northern Ireland. J. Quatern. Sci., 14, 45–
57.
Laberg, J.S., and T.O. Vorren (2000): The Trænadjupet Slide, offshore Norway—
morphology, evacuation and triggering mechanisms. Mar. Geol., 171, 95–114.
Long, D., D.E. Smith, and A.G. Dawson (1989): A Holocene tsunami deposit in
eastern Scotland. J. Quatern. Sci., 4, 61–66.

366 D. Long and R. Holmes
Ringrose, P.S. (1989): Recent fault movement and palaeoseismicity in western Scot-
land. Tectonophysics, 163, 305–314.