Slope failure dynamics and impacts from seafloor and shallow sub-seafloor geophysical data: case studies from the COSTA project
Introduction
A tremendous effort has been made in the last few years to characterize and better understand seafloor failures in European margins (Mienert and Weaver, 2002 and references therein) and elsewhere (Locat and Mienert, 2003 and references therein). The more that is known about continental margins, the clearer it becomes that submarine slides are a widespread phenomenon (Canals, 1985, Hühnerbach et al., 2004). The interests of the oil industry have triggered their study, jointly with other deepwater geohazards, mostly during the last decade as related to the exploration and exploitation of hydrocarbon resources in the deep sea (Campbell, 1999). A variety of slides, often related to fluid escape, is known to occur in the most important offshore oil provinces such as the Norwegian margin, the Gulf of Guinea, the Gulf of Mexico and the Caspian Sea (Barley, 1999). The second largest gas field discovery off Norway, the Ormen Lange field, is located within the scar created by the Storegga Slide, possibly the largest submarine slide in the world ocean (Bryn et al., 2003a).
Seafloor failures represent a major threat not only to the oil and offshore industries but also to the marine environment and coastal facilities. It is well known that large historical seafloor failures have engendered destructive tsunamis. Recent results indicate that the large tsunamis that devastated Lisbon and struck the Gulf of Cadiz and North Atlantic coasts both in Europe and Africa in 1755 following a magnitude ∼8.5 earthquake probably had a landslide contribution (Gracia et al., 2003). Seismicity in the Southwestern Iberian margin results from tectonic activity along the Europe–Africa plate boundary connecting the Azores Triple Junction to the west to the Gibraltar Strait to the east (Zitellini et al., 2001). The Lisbon event, as it is known, represents the largest natural catastrophe in Western Europe since the Roman period, which resulted in about 60,000 casualties in Portugal alone (Baptista et al., 1998). The destruction of Lisbon, at the time one of the main capitals in Europe, terrified European society. At present, there is an important international on-going effort off Portugal to further investigate the source area of the Lisbon earthquake and related submarine landsliding (Zitellini et al., 2001, Gracia et al., 2003).
The breaking of submarine telegraph cables during the Grand Banks event following an earthquake in 1929 is also an outstanding case that had a profound impact on deep sea sedimentological research. The water depth of the area affected ranges from 650 to about 2800 m, and the distance between the scar rim and the most distal deposit is >850 km. It appears that the Grand Banks mass movement may have reached a maximum velocity of about 70 km/h according to Heezen and Ewing (1952). The thickness of the turbidity current was of the order of hundreds of meters as determined from erosional trimlines (Piper and Aksu, 1987). Detailed descriptions of the Grand Banks event and resulting deposits can be found in Rupke (1978) and Piper et al. (1999). This event occurred at a time when the now widely accepted concepts of turbidity currents and the continuum of submarine mass gravity flows (from slumps to debris flows to turbidity currents) had not yet been conceived. Heezen and Ewing (1952) and Heezen and Hollister (1971) shook the scientific community after convincingly identifying the Grand Banks slumps and turbidity current as the cause of cable breaks south of Newfoundland in 1929. Their works followed famous earlier papers by Kuenen (1937) and Kuenen and Migliorini (1950) where these authors demonstrated the existence of turbidity currents and showed some of their properties after conducting a series of classic flume experiments. Other key pioneer papers that greatly helped in establishing the current background on mass gravity flows were those of Morgenstern (1967), Hampton (1972) and Middleton and Hampton (1976) to cite just a few. As correctly pointed out by Rupke (1978), these theories revolutionized the study of clastic sediments and enormously stimulated research on deep-sea sedimentary processes.
Now, we know that sediment failure around the epicenter of the 1929 Grand Banks earthquake shows a downslope transition from retrogressive thin-skinned rotational slumps, through debris flows, to erosional features cut by turbidity currents, to turbidite deposits (Piper et al., 1999). The 1929 turbidity current was thus triggered by prolonged numerous relatively small failures nourishing it over a period of about 11 h (Hughes Clarke, 1988). While limited deep-towed side scan sonar imagery, high resolution seismic reflection profiles, sediment cores, in situ shallow geotechnical measurements and submersible observations are available (Hughes Clarke et al., 1989, Piper et al., 1985, Piper et al., 1999), multibeam mapping of the continental slope area disturbed by the 1929 Grand Banks earthquake has not been completed, a surprising situation. The need for new data including swath bathymetry has steered an international consortium of research teams, which has advanced plans to deploy there the best geophysical tools available for deep seafloor and sub-seafloor imaging. The benefits from such an endeavor are anticipated to be of major importance.
A third recent submarine landslide that had a major impact both on coastal facilities and on the scientific community in Europe occurred off the French town of Nice in the Northwestern Mediterranean the 16th of October 1979. The source area was the prograding prodelta of the Var River that accumulated on a very narrow shelf. The nearby Monaco observatory registered no earthquake that could have triggered the slide (Malinverno et al., 1988). Because of the very steep nature of the seafloor off Nice, undercutting cannot be excluded as a concurrent potential triggering mechanism. In addition, sediment failures off Nice are favored by the common occurrence of underconsolidated, meter-thick sediment layers (Cochonat et al., 1993, Klaucke and Cochonat, 1999) although ridge-forming normally consolidated to overconsolidated sediments could also be involved (Mulder et al., 1993, Mulder et al., 1994). Three types of sediment failure have been distinguished by Klaucke and Cochonat (1999): superficial slumping, deep-seated failure often associated with successive rotational slides and gullying of the canyon walls.
The shelf and upper slope 1979 slide evolved into a turbidity current, which, as in the Grand Banks case, broke submarine communication cables. The calculated peak velocity of the mass movement was 40 km/h according to Gennesseaux et al. (1980). The suction effect of the downslope-moving sediment mass generated first a retreat of the sea and, second, a several meters high tsunami wave (Groupe ESCYANICE, 1982, Malinverno et al., 1988). As a consequence of the event, part of a land filled area reclaimed to the sea to enlarge the airport of Nice was destroyed, bulldozers were dragged deep into the sea and various people were killed (Savoye, 1991, Mulder et al., 1997). In addition to the work already carried out, the stability of the Nice offshore area is being actively investigated with a priority for observations with highly capable imaging tools, in situ measurements, laboratory tests and modeling (Savoye et al., 2004, Sultan et al., 2004). New in situ instruments such as IFREMER's flexible penetrometer (Penfeld) have been first deployed off Nice.
In a date as recent as July 1998, a tsunami most probably generated by a submarine slump hit the Sissano coast in northwestern Papua-New Guinea (Tappin et al., 1999). Wave heights of 10 m were observed along a 25-km stretch of coastline with maximum heights of 15 m and overland flow velocities of 54–72 km/h. The death toll was over 2200, surpassed in the XXth century only by a tsunami on the coast of Sanriku, Japan, in 1933 (Kawata et al., 1999). The tsunamigenic submarine slump occurred 25 km offshore and was itself probably triggered by an estimated 7.0 magnitude earthquake. The Sissano tsunami is the first that has been comprehensively investigated very soon after its occurrence by seabed and sub-seabed imaging, sediment coring, ROV and manned submersible observations, measurements of potential fields and computer simulations. The approximately 760-m-thick, 5–20-km3 slump took place in an arcuate, amphitheatre-shaped structure made of fine grained, cohesive and stiff sediments that failed by rotational faulting. Fissures, brecciated angular sediment blocks, vertical slopes, talus deposits and evidence of active fluid expulsion have been found in the amphitheatre area. A failure plane with at most a 100-m high exposed scar has been identified on the slump headwall. Also, the occurrence of several events of different ages in the same source area has been postulated. Local seabed morphology resulted in focusing the magnitude and wave-height distributions of the tsunami along the coast (Tappin et al., 2001).
The most recent submarine landslide generating a tsunami that we are aware of, took place on the flanks of the volcanic island of Stromboli, Thyrrhenian Sea, while writing the present paper (December 30, 2002). According to an oral account by S. Tinti from the University of Bologna, Italy, two successive slides, one subaerial and submarine and the other only subaerial, affected an area prone to instability known as Sciara del Fouco. The total volume of rock and debris remobilised was about 28.5 millions of m3 (Bosman et al., 2004). The first slide was responsible for the observed tsunami, which flooded part of the lowlands to the north of the island. The observed height of the wave was up to 10 m at specific locations. There were no casualties. The Stromboli tsunami wave was recorded by tide gauges in nearby islands and also in Milazzo, north of Sicily, where tankers were displaced during oil transfer operations, and oil depots on the coast were close from being hit by the wave.
The above accounts only represent a small part of all the known occurrences of submarine slides in historical times. To illustrate our points, we have deliberately chosen a few slides that generated tsunamis since these are the ones that have a stronger social, economical and scientific impact. Many other submarine landslides are known to have occurred not only during the historical epoch but also throughout the Holocene (Canals, 1985, Hühnerbach et al., 2004). Note that submarine landslides, eventually associated with tsunamis, might be rather frequent along European and North Atlantic margins, even on segments that can be considered tectonically quiet (i.e., Lisbon, Grand Banks and Nice slides and tsunamis).
One of the major advantages of studying geologically recent or historical seafloor mass movements is that they can be much better constrained than older events in terms of resulting morphologies, deposits, dynamics, impacts and ages. To achieve such knowledge, state-of-the-art high resolution geophysical tools (i.e., swath bathymetry systems, deep-towed side scan sonars, high to ultra-high resolution 2D and 3D seismic reflection profiling) are required to provide seafloor and sub-seafloor images of unprecedented quality that can then be used to investigate the above points. The enormous improvement in surveying equipment during the last few years is bringing to the surface events and impacts, jointly with their fine-grained details, that could not be resolved previously. Coring is a necessary complement to get datable samples for events that have occurred in pre-historical times or whose timing is not well known even if historical.
One of the main tasks within the “Continental Slope Stability” (COSTA) project has been to investigate slope failure dynamics and impacts from seafloor and sub-seafloor shallow geophysical data with the aim to assess:
- –
External morphology and internal structure of slope failures and resulting deposits
- –
Slip plane geometries for small, medium and megaslide events
- –
Run-out distances and flow pathways
- –
Triggering mechanisms
- –
Ages of slide events, either single-phased or multi-phased, and recurrence intervals
To achieve the above aims, which overall could illustrate the variability of submarine sediment failures, the research effort focused on eight pre-Holocene to present case studies representing the variety of submarine instabilities that can be found along ocean margins. Seafloor instability events in this paper are now among the best studied in the world. Describing the main results achieved through their study, extracting overall conclusions and distillating implications are the primary goals of the current paper, which also includes a review and summary of previously published data.
Section snippets
Setting of the studied slides
Six of the slides studied are located in Europe's margin and have been systematically and intensively investigated within the COSTA project. These are from north to south Traenadjupet, Storegga and Finneidfjord Slides, off Norway, Afen Slide from the Faeroe-Shetland Channel, and BIG'95 Slide and the Central Adriatic sediment deformation belt from the Mediterranean Sea (Fig. 1). The Canary Slide affecting an ocean island flank has been added as an end member not represented by the European case
Methods
High resolution state-of-the-art geophysical methods were applied to the case studies considered within the COSTA project. These include both seafloor and shallow sub-seafloor imaging tools deployed both near the sea surface and near the ocean bottom. Sediment cores provided materials to groundtruth geophysical interpretations and to perform age analyses of the events. Table 2 summarizes the methods used to investigate the slope failure dynamics and impacts of each of the eight landslides of
Mass movement types
Before assigning a specific mass movement type to any of the COSTA instabilities, it is convenient to introduce a word of caution about terminology. There is certainly a terminology problem in the literature since some of the terms (i.e., slide and slump) are often used in a loose way to refer to almost any type of submarine mass movement rather than reserving them for specific mass movement types following precise definitions (Canals, 1985). Indeed, the readers will have noted that the term
Slide dimensions
The slides studied within the COSTA project are variable in size, from the giant Storegga Slide to the comparatively tiny Finneidfjord Slide (Fig. 8 and Table 5). The largest slides (Storegga, Canary and Traenadjupet) develop off open slopes, while those on the margins of semi-enclosed basins are smaller and show changes in direction reflecting topographic control on sediment flow (BIG'95 and Gebra Slides) (Fig. 8). Whereas calculating the approximate area of the landslides (scar and
Scars and slip planes
Scars are perhaps the most critical area to be investigated within submarine instabilities since they represent the source area and allow comparison with nearby non-failed areas. To a large extent, scar location, morphology and size, jointly with the nature of the failed material and the triggering mechanism, influence the type of mass movement, its dynamics and the volume of the resulting deposit. Of paramount importance are also the seafloor dip and the inclination of slip planes at and
Run-out distances
Run-out could be defined as the horizontal distance between the upper edge of the slide headwall and the distalmost point reached by sediments mobilised during a slide event. In multi-staged slides, run-out tends to increase with time because of upslope retrogression, continued transport and downslope extension of the released sediments. Different slide phases usually have different run-outs, with the largest failures in terms of volume having in general the largest run-outs. Often, secondary
Triggering mechanisms
A triggering mechanism, or a combination of triggering mechanisms is required to destabilise sedimentary packages already prone to failure because of a set of preconditioning factors. Failure occurs when the downslope oriented shear stress (driving shear stress) exceeds the shear strength (resisting stress) of the material forming the slope, as expressed by the well known Mohr–Coulomb failure criterion:where τf is the shear strength (equivalent to the shear stress at failure), c
Ages of slide events
The available information about the ages of the submarine landslides studied within the COSTA project is summarized in Table 8, where information about dating techniques, number of datings, sedimentary units dated and, in some cases, recurrence intervals is provided too. It is best reading Table 8 jointly with Table 2, Table 3 where information on the number and type of sediment cores, and on the retrogressive and/or multi-staged character for each of the landslides can be found. In any case,
Concluding remarks
The COSTA project has provided a unique opportunity to investigate a variety of geologically recent submarine instabilities along the northern and southern continental margins of Europe. The effort performed and the results achieved make clear that a combination of regional scale studies with thorough analyses of individual instabilities from various points of view is essential to progress towards a comprehensive understanding of mass wasting events at a global scale. As important as these
Acknowledgements
This study was supported by “Continental Slope Stability” (COSTA, ref. EVK3-1999-00028) project, with contributions from the following EU and Spanish research projects: “European Deep Ocean Margins” Research Training Network (EURODOM, ref. RTN2-2001-00281), “European margin strata formation” (EUROSTRATAFORM, ref. EVK3-2001-00200), “Procesos de inestabilidad sedimentaria en márgenes continentales e insulares españoles: los megadeslizamientos del Ebro y de Canarias” (GRANDES, ref. MAR98-0347),
References (270)
- et al.
Determination of sediment volumes, accumulation rates and turbidite emplacement frequencies on the Madeira Abyssal Plain (NE Atlantic): a correlation between seismic and borehole data
Mar. Geol.
(1999) - et al.
Late Quaternary sedimentation patterns of the Ebro turbidite systems (Northwestern Mediterranean): two styles of deep-sea deposition
Mar. Geol.
(1990) - et al.
Sedimentary evolution of the Pliocene and Pleistocene Ebro margin, northeastern Spain
Mar. Geol.
(1990) - et al.
Constrains on the source of the 1755 Lisbon tsunami inferred from numerical modelling of historical data
J. Geodyn.
(1998) - et al.
Inferred gas hydrates and clay diapirs near the Storegga Slide on the southern edge of the Voring Plateau, offshore Norway
Mar. Geol.
(2000) - et al.
Geological controls on the Storegga gas-hydrate system of the mid-Norwegian continental margin
Earth Planet. Sci. Lett.
(2003) - et al.
Tephra layers in late Quaternary sediments of the central Adriatic Sea
Mar. Geol.
(1998) A discussion and classification of subaqueous mass-transport with particular application to grain flow and fluxoturbidites
Earth-Sci. Rev.
(1975)- et al.
The late Holocene Gargano subaqueous delta, Adriatic shelf: sediment pathways and supply fluctuations
Mar. Geol.
(2003) - et al.
Gas seeps and rock formation in the northern Adriatic Sea
Cont. Shelf Res.
(2002)
Echo character of the Norwegian–Greenland Sea: relationship to Quaternary sedimentation
Mar. Geol.
The Storegga Slides: evidence from eastern Scotland for a possible tsunami
Mar. Geol.
Flow models of small to medium-scale debris flows originating from compacted clay materials
Mar. Geol.
Submarine slope stability on high-latitude glaciated Svalbard–Barents Sea margin
Mar. Geol.
Faulting in the Madeira Abyssal Plain sediments
Mar. Geol.
The upper Cainozoic of the Norwegian continental shelf correlated with the deep sea record of the Norwegian Sea and the North Atlantic
Mar. Pet. Geol.
Paleoceanographic implications of the Messinian surface in the Valencia Trough, northwestern Mediterranean Sea
Evidence for long-term instability in the Storegga Slide region off western Norway
Mar. Geol.
The Ebro continental shelf: Quaternary seismic stratigraphy and growth patterns
Mar. Geol.
Evidence for the multi-stage formation of the southwestern Valencia Trough
Mar. Pet. Geol.
Accumulation rates and 137Cs distribution in sediments off the Po River delta and the Emilia-Romagna coast (nothwestern Adriatic Sea, Italy)
Cont. Shelf Res.
On the post-25 Ma geodynamic evolution of the Western Mediterranean
Tectonophysics
Continental shelf glacial troughs and continental rise development off the North Antarctic Peninsula
Quat. Sci. Rev.
La edad del volcanismo de las islas Columbrete Grande y Alborán (Mediterráneo occidental)
Bol. Geol. Min.
Foreland deformational pattern in the southern Adriatic Sea
Ann. Geofis.
Short-term climate changes during the Last Glacial–Holocene transition: comparison between Mediterranean records and the GRIP event stratigraphy
J. Quat. Sci.
Practical Seismic Interpretation
Seismic facies investigation of the late Quaternary glacial history of Bransfield Basin
Record of Holocene glacial oscillations in Bransfield Basin as revealed by microfossil assemblages
Antarct. Sci.
Deepwater problems around the world
Lead. Edge
Pliocene–Quaternary sedimentation in the Northern Apennine Foredeep and related denudation
Geol. Mag.
Glossary of Geology
Polygonal fault systems on the mid-Norwegian margin: a long-term source for fluid flow
The role of free gas in the activation of submarine slides in Finneidfjord
Handbook of Seafloor Sonar Imagery, Wiley-Praxis Ser. in Remote Sensing
The Storegga tsunami along the Norwegian coast, its age and runup
Boreas
Hydrothermal activity at Hook Ridge in the Central Bransfield Basin, Antarctica
Geo Mar. Lett.
Fast evolution of a submarine volcanic flank experiencing a large-scale landslide: the case of Stromboli, Aeolian Islands
Rapp. Comm. Int. Mer. Médit.
Cyclic behavior of Eel River sediments: a possible explanation for the paucity of submarine landslide features
AGU, EOS Abstr.
How we conclude on the Storegga Slide risk, COSTA
The Storegga Slide Complex: repeated large scale sliding in response to climatic cyclicity
Submarine slides on the Norwegian continental margin, with special emphasis on the Storegga area
Publ. Cont. Shelf Inst.
A giant three-stage submarine slide off Norway
Geo Mar. Lett.
The storegga slide
Philos. Trans. R. Soc. London Ser. A: Math. Phys. Sci.
Imaging the Afen Slide from commercial 3D seismic—methodology and comparison with high-resolution data
Images of the seabed in the Faeroe-Shetland Channel from commercial 3D seismic data
Mar. Geophys. Res.
Deepwater geohazards: how significant are they?
Lead. Edge
Cited by (385)
Free fall penetration test the undrained shear strength of submarine sediments
2024, Measurement: Journal of the International Measurement ConfederationPredicting impact forces on pipelines from deep-sea fluidized slides: A comprehensive review of key factors
2024, International Journal of Mining Science and TechnologyCoupled stability of offshore wind monopile foundations and submarine slopes under periodic loads
2023, Applied Ocean ResearchSand-rich Pleistocene deep-water channels and their implications for gas hydrate accumulation: Evidence from the Qiongdongnan Basin, northern South China Sea
2023, Deep-Sea Research Part I: Oceanographic Research Papers