Elsevier

Marine Geology

Volumes 307–310, 15 April 2012, Pages 1-21
Marine Geology

Failure mechanisms of Ana Slide from geotechnical evidence, Eivissa Channel, Western Mediterranean Sea

https://doi.org/10.1016/j.margeo.2012.02.010Get rights and content

Abstract

This work deals with the failure mechanisms of Ana Slide in the Eivissa Channel, in between the Iberian Peninsula and the Balearic Islands, under the effects of gas charging and seismic loading. In situ geotechnical tests and sediment cores obtained at the eastern Balearic slope of the Eivissa Channel suggest that the basal failure surface (BFS) developed as a result of subtle contrasting hydro-mechanical properties at the boundary between a fine-grained unit (U6) overlying a methane-charged relatively coarser unit (U7). Past methane seepage is inferred from seismic reflection profiles and high magnetic susceptibility values in sediments from the slide headwall area. Past methane charging is also supported by further seismic reflection data and isotopic analyses of benthic foraminifera published separately. The possibility of failure for different critical failure surfaces has been investigated by using the SAMU-3D slope stability model software taking into account the role of free methane in the development of the landslide. Failure would occur after SAMU-3D if the undrained shear strength of units U6 and U7 is strongly degraded (i.e. 95%). Wheeler's theory suggests that a 9% free gas saturation would be required to reduce the undrained shear strength by 95%. However, the theory of the undrained equilibrium behaviour of gassy sediments for this methane concentration shows that the excess fluid pressure generated by gas exsolution, estimated at 12% of the effective stress, is not high enough to bring the slope to fail. This led us to consider seismic loading as an additional potential failure mechanism despite the lack of historical data (including instrumental records) on seismicity in the Balearic Islands, therefore assuming that the historical period is not necessarily representative of seismic activity further back in time (i.e. when Ana Slide occurred ~ 61.5 ka ago). Considering current slope conditions, the most critical failure surface obtained by SAMU-3D relates to peak ground accelerations (PGA) of 0.24 g, which relates to magnitude moment Mw = 5 at epicentral distances of 1 km, and 7  Mw  5 at epicentral distances ≤ 15 km to Ana Slide. However, no active faults have been identified at so short distance from Ana Slide. Only when shear strength is degraded due to the presence of free methane in units U6 and U7 is considered, the most critical failure surface obtained by SAMU-3D fits with lower magnitude and larger epicentral distances. Consequently, the most plausible hypothesis to explain the occurrence of Ana Slide is the combination of free gas and seismic loading.

Highlights

► We address the role of methane seepage and seismic loading on a submarine landslide. ► Past methane seepage is inferred from seismic data and magnetic susceptibilities. ► Slope failure due to overpressure by gas expansion is discarded. ► Undrained shear strength degradation due to bubble pressure can explain the failure. ► Seismicity alone can hardly have been the triggering without the presence of gas.

Introduction

Ana Slide covers 6 km2 and is located on the eastern Balearic flank of Eivissa Channel (Fig. 1a). This small landslide is aligned with three other landslides (Joan, Nuna and Jersi, Fig. 1b) along the 0°48′E meridian at water depths ranging from 600 m depth at the rim of the shallowest headscarp (Joan Slide) down to 900 m at the lower edge of the deepest landslide toe (Jersi Slide) (Lastras et al., 2004, Lastras et al., 2007). The current seafloor expression of Ana Slide ranges from 635 m to 790 m of water depth, with an average slope of 1.6° (Lastras et al., 2004) (Fig. 1c). Previous work based on multibeam bathymetry, backscatter and high-resolution seismic reflection profiles indicated that modest mass transfer accompanied the downslope propagation of the deformation front of Ana Slide, which extensively remoulded the underlying slope sediments without necessarily displacing them too far downslope. Only a slight displacement is observed for most of the sediment in the central and lower sections of the landslide (Lastras et al., 2004). Extensional ridges in the headwall area of Ana Slide correspond to detached subvertical slabs of partially disturbed sediments, possibly indicating a retrogressive character (Lastras et al., 2006). The observation that the shear planes of Ana, Joan, Nuna and Jersi landslides correspond to the same seismic reflector (Lastras et al., 2004) points to a likely stratigraphic control of slope stability along the eastern slope of the Eivissa Channel and to common causative factors and triggering mechanisms.

Pockmarks near the headwall scarps of Nuna and Ana slides suggest that fluid escape processes took place in the vicinity of the slides (Lastras et al., 2006, Lastras et al., 2007). Fluid escape structures of the Balearic Promontory have been described as extrusion features by gas of thermogenic origin related to Plio-Quaternary submarine volcanism and faulting (Acosta et al., 2001a, Acosta et al., 2001b, Maillard and Mauffret, in press). Seismic data acquired across Ana Slide (see Fig. 2 in Berndt et al., 2012) show high amplitude anomalies below the failed sediment mass, which are attributed to free gas, even though the polarity of these reflections is ambiguous (Berndt et al., 2012). The presence of methane in the recent past has been also inferred from δ13C analyses on benthic foraminifera Hyalinea balthica and Uvigerina peregrina assemblages from sediment samples extracted from a sediment core recovered at Ana's headwall (Panieri et al., 2012).

The role of gas charging in slope instability is generally related to gas exsolution and overpressure formation during sea-level falls. Sobkowicz and Morgenstern (1984) consider that if equilibrium conditions prevail during undrained unloading (equivalent to sea-level falls) of gassy sediments (i.e. sediments containing large gas bubbles), the pore pressures remain close to the liquid/gas saturation pressure and effective stresses are reduced significantly. Pore pressure response due to changes in total stress such as sea-level falls generates gas exsolution and subsequent loss of effective stress, which leads to a reduction in the strength of the sediment. Other models developed to predict the behaviour of gassy sediments assume that the water phase is continuous, thus remaining valid the principle of effective stress (Wheeler, 1988 and Grozic et al., 2005). According to Wheeler (1988), a lower bound for the undrained shear strength is defined during undrained failure, which allows quantifying the degree of undrained shear strength degradation due to the presence of gas.

In this article we present the geotechnical properties of the upper 28 m of Ana Slide sedimentary succession based on in situ piezocone tests and laboratory analyses and additional tests of sediment cores. The aim of our study is to analyse the failure mechanisms potentially involved in Ana Slide occurrence. To achieve our aim we use the SAMU-3D stability analysis model taking into account the role of (methane) gas charging and seismic loading, for which the geological context is also considered. The role of methane charging as the main causative factor of Ana Slide and, likely, of the neighbouring slope failures in the Eivissa Channel, is evaluated by using Sobkowicz and Morgenstern (1984) and Wheeler (1988) theories.

Section snippets

Geological setting

The Eivissa Channel occupies the western end of the Balearic Promontory, which separates the Valencia Trough to the north from the Algero-Balearic Basin to the south. The Balearic Promontory consists of the Menorca–Mallorca and the Eivissa–Formentera structural blocks, classically viewed as an extension of the external zone of the Betic Ranges in the Iberian Peninsula (Maillard and Mauffret, 1999).

The current physiography of the Balearic Promontory results from Tertiary and older tectonic

Background data

The Eivissa Channel was surveyed in 1995 and 2002 during BIG'95 and MARINADA cruises onboard the research vessel BIO Hespérides, during which swath bathymetry data and TOPAS very-high resolution seismic reflection profiles were acquired (see compilation in Lastras et al., 2004). Side-scan sonar data were collected with the MAK-1M system during the BALICAT survey in 2004 onboard the R/V Professor Logachev (Lastras et al., 2006). Chirp-sonar and sparker seismic reflection profiles and gravity

Intact sedimentary succession

The geotechnical stratigraphy derived from the corrected cone resistance (qt) and the sleeve friction (fs) profiles was established and correlated with seismic reflectors in TOPAS M-3 profile (Fig. 2a). It consists of a top Post-Slide unit (PS) overlying 14 units that slightly thicken downslope (Figs. 2b and 3). The PS unit is rather uniform, with similar qt trend and thickness (~ 2.5 m) at all CPTU sites except site PFM-05S01 in which it is ~ 3 m (Fig. 3). Sharp peaks in u2 profiles at variable

Evidence of past methane seepage

In this section we describe, along a downslope transect of sediment cores, MS profiles that in some cases could be interpreted as an indication of past methane seepage. Core KS18, retrieved from the non-destabilised slope off the upper slide headwall rim, displays low MS values in units PS, U1 and U2 (10–15 · 10 5 SI/volume units, Fig. 5a), which are consistent with MS measurements in sediment cores recovered at more than 2800 m of water depth in the Balearic Abyssal Plain east of the Balearic

3D-stability back-analysis

We have first reconstructed the pre-slide bathymetry by subtracting the top 2.5 m that we consider representative of the thickness of PS unit and adding the relevant sediment packages where overburden removal is observed (Fig. 14a, b). The layer model required for SAMU-3D is a simplification of the sedimentary architecture shown in Fig. 2b. Since the BFS is at the U6–U7 boundary (or close to it), our layer model consists of 7 layers that represent units U1 to U7 and a 20 m thick basement

Preconditioning factors

The results obtained from the slope stability analysis using the SAMU-3D software demonstrate that the Ana Slide failure (FOS ≤ 1) can be explained by: (i) the presence of gas (methane) in the sediment (9% in unit U7), (ii) seismic loading resulting from PGA ≥ 0.24 g and (iii) the combination of the presence of methane in the sediment with lower concentrations (in unit U7) and seismic loading resulting from 0.06 g  PGA < 0.24. This shows that the accumulation of free methane in a particular

Conclusions

This study investigates the failure mechanisms of Ana Slide, a 6 km2 shallow submarine landslide located between 635 and 790 m of water depth in the slope west of the Balearic Promontory within the Eivissa Channel, with an estimated age of ~ 61.5 ka. The basal failure surface (BFS) developed between two units named U7 (below) and U6 (above). Unit U6 is interpreted to be finer than U7, which appears coarser at its base. Subtle contrasting hydro-mechanical properties are found at the U7–U6 boundary.

List of symbols

    BP

    before present

    Bq

    pore pressure ratio

    c

    effective cohesion, kPa

    d

    volume fraction of gas bubbles

    CU

    consolidated-undrained

    FOS

    factor of safety

    fs

    sleeve friction, kPa

    kyr

    kilo year, used for time intervals

    ka

    kilo anni, used for absolute age

    MS

    magnetic susceptibility

    Mw

    magnitude moment

    Nσt

    cone parameter = (qt  σv0)/σp

    Nkt

    cone factor

    OCR

    overconsolidation ratio

    p

    mean effective stress (in triaxial tests), kPa

    PGA

    peak ground acceleration, m∙s 2

    q

    deviatoric stress (in triaxial tests), kPa

    qc

    cone tip resistance, kPa

    qt

Acknowledgements

We thank IFREMER team and crew of R/V L'Atalante for their support during the PRISME 2007 cruise. This work has been co-founded by IFREMER and the CONSOLIDER 2010-GRACCIE project (Ref. CSD2007-00067), the EC HERMIONE project (contract no 226354-HERMIONE), and a Generalitat de Catalunya grant for excellence research groups to GRC Geociències Marines (Ref. 2009 SGR 1305). The publication reflects only the views of the authors. The EC is not liable for any use that may be made of the information

References (63)

  • B.A.A. Hoogakker et al.

    Variations in terrigenous dilution in western Mediterranean Sea pelagic sediments in response to climate change during the last glacial cycle

    Marine Geology

    (2004)
  • T.J. Kvalstad et al.

    Soil conditions and slope stability in the Ormen Lange area

    Marine and Petroleum Geology

    (2005)
  • J.C. Larrasoaña et al.

    Magnetic susceptibility of eastern Mediterranean marine sediments as a proxy for Saharan dust supply?

    Marine Geology

    (2008)
  • M.W. Lee

    Elastic velocities of partially gas-saturated unconsolidated sediments

    Marine and Petroleum Geology

    (2004)
  • M.W. Lee et al.

    Unique problems associated with seismic analysis of partially gas-saturated unconsolidated sediments

    Marine and Petroleum Geology

    (2009)
  • D. Leynaud et al.

    3-D slope stability analysis: a probability approach applied to the Nice slope (SE France)

    Marine Geology

    (2010)
  • A. Maldonado et al.

    Depositional patterns and Late Quaternary evolution of two Mediterranean submarine fans: a comparison

    Marine Geology

    (1979)
  • C. Olivera et al.

    Seismicity of the Valencia trough and surrounding areas

    Tectonophysics

    (1992)
  • G. Panieri et al.

    Tracing seafloor methane emissions with benthic foraminifera: results from the Ana submarine Slide (Eivissa Channel, Western Mediterranean Sea)

    Marine Geology

    (2012)
  • S.G. Robinson

    The late Pleistocene paleoclimatic record of North Atlantic deep-sea sediments revealed by mineral-magnetic measurements

    Physics of the Earth and Planetary Interiors

    (1986)
  • N. Sultan et al.

    Potential role of compressional structures in generating submarine slope failures in the Niger Delta

    Marine Geology

    (2007)
  • B.E. van Dongen et al.

    Formation iron sulfide nodules during anaerobic oxidation of methane

    Geochimica et Cosmochimica Acta

    (2007)
  • C. Waelbroeck et al.

    Sea-level and deep water temperature changes derived from benthic foraminifera isotopic records

    Quaternary Science Reviews

    (2002)
  • J. Acosta et al.

    Pockmarks in the Ibiza Channel and western end of the Balearic Promontory (western Mediterranean) revealed by multibeam mapping

    Geo-Marine Letters

    (2001)
  • J. Acosta et al.

    The Balearic Promontory geomorphology (western Mediterranean): morphostructure and active processes

    Geomorphology

    (2002)
  • J.-F. Bourillet et al.

    Behaviour of a piston corer from accelerometers and new insights on quality of the recovery

  • A. Cattaneo et al.

    Age constraints and sediment properties of Ana Slide (Balearic Sea, Western Mediterranean) and implications on age dating of submarine landslides

    Geophysical Research Abstracts

    (2011)
  • Z. Chen et al.

    3-dimensional slope stability analysis using non-associative stress–strain relationships

    Science China Technological Sciences (Science in China Series E)

    (2009)
  • R.F. Craig

    Craig's Soil Mechanics

    (2005)
  • G. Dan et al.

    Quantifying the role of sandy–silty sediments in generating slope failures during earthquakes: example from the Algerian margin

    International Journal of Earth Sciences

    (2009)
  • J.A. Dean

    Lange's Handbook of Chemistry

    (1999)
  • Cited by (32)

    • A new methodology to assess the potential of conjectural trigger mechanisms of submarine landslides exemplified by marine gas occurrence on the Balearic Promontory

      2021, Engineering Geology
      Citation Excerpt :

      The latter occurred approximately 60 ka. ago, is 4 km long from the headwall scar to the frontal lobe and has been thoroughly investigated by geophysical and geotechnical means (Lastras et al., 2004; Lastras et al., 2006; Berndt et al., 2012; Lafuerza et al., 2012; Panieri et al., 2012). The seismic investigations by Berndt et al. (2012) show indication for a long-term, locally established fluid migration system.

    • Apparent overconsolidation and its implications for submarine landslides

      2020, Engineering Geology
      Citation Excerpt :

      The possible causes of AOC in marine sediments have been narrowed to the following three factors: Inorganic cementation by agents such as silicates, carbonates, iron oxides or aluminum oxide (interparticle bonding through chemical cementation) (Anandarajah and Lavoie, 2002; Brandes et al., 1996; Bryant and Bennet, 1988; Eng, 2004; Hamilton, 1964; Keller and Bennett, 1973; Lafuerza et al., 2012; Richards and Hamilton, 1967; Silva and Jordan, 1984; Veyera et al., 2001). This explanation has been used most often in the scientific literature, since it effectively explains the AOC observed in some areas and specific samples.

    View all citing articles on Scopus
    View full text