Geotechnical Characteristics and Slope Stability

Analysis on the Deeper Slope of the Ligurian Margin,

Southern France

Fei Ai, Annika Förster, Sylvia Stegmann, and Achim Kopf

Abstract
Submarine slope failures of various types and sizes are common along the tectonic and
seismically active Ligurian margin, northwestern Mediterranean Sea, primarily because of
seismicity up to ~M6, rapid sediment deposition in the Var fluvial system, and steepness of
the continental slope (average 11 ). We present geophysical, sedimentological and
geotechnical results of two distinct slides in water depth >1,500 m: one located on the
flank of the Upper Var Valley called Western Slide (WS), another located at the base of
continental slope called Eastern Slide (ES). WS is a superficial slide characterized by a
slope angle of ~4.6 and shallow scar (~30 m) whereas ES is a deep-seated slide with a
lower slope angle (~3 ) and deep scar (~100 m). Both areas mainly comprise clayey silt
with intermediate plasticity, low water content (30–75 %) and underconsolidation to strong
overconsolidation. Upslope undeformed sediments have low undrained shear strength
(0–20 kPa) increasing gradually with depth, whereas an abrupt increase in strength up to
200 kPa occurs at a depth of ~3.6 m in the headwall of WS and ~1.0 m in the headwall of
ES. These boundaries are interpreted as earlier failure planes that have been covered by
hemipelagite or talus from upslope after landslide emplacement.
Infinite slope stability analyses indicate both sites are stable under static conditions;
however, slope failure may occur in undrained earthquake condition. Peak earthquake
acceleration from 0.09 g on WS and 0.12 g on ES, i.e. M5–5.3 earthquakes on the spot,
would be required to induce slope instability. Different failure styles include rapid
sedimentation on steep canyon flanks with undercutting causing superficial slides in the
west and an earthquake on the adjacent Marcel fault to trigger a deep-seated slide in the east.

F. Ai (*)
MARUM-Center for Marine Environmental Sciences, University of Bremen, Leobener Straße, 28359 Bremen, Germany
Faculty of Geosciences, University of Bremen, Leobener Straße, 28359 Bremen, Germany
State Key Laboratory of Geomechanics and Geotechnical Engineering, Institute of Rock and Soil Mechanics, Chinese Academy of Sciences,
430071 Wuhan, China
e-mail: [email protected]
A. Förster
Institute of Geosciences and Geography, Martin-Luther-University Halle-Wittenberg, von Seckendorff Platz 3, 06120 Halle (Saale), Germany
e-mail: [email protected]
S. Stegmann  A. Kopf
MARUM-Center for Marine Environmental Sciences, University of Bremen, Leobener Straße, 28359 Bremen, Germany
Faculty of Geosciences, University of Bremen, Leobener Straße, 28359 Bremen, Germany
e-mail: [email protected]; [email protected]

K. Sassa et al. (eds.), Landslide Science for a Safer Geoenvironment, Vol. 3, 549
DOI 10.1007/978-3-319-04996-0_84, # Springer International Publishing Switzerland 2014
550 F. Ai et al.

Keywords
Submarine slope failure  Geotechnical characteristics  Slope stability analysis  Ligurian
margin

Introduction involves moderate seismic activity with common earthquake

magnitudes of M2.2 to M4.5 (Fig. 1a). However, earthquake
Submarine slope failures represent the main agents of sedi- magnitudes up to M6.8 (e.g. 1887 Ligurian earthquake) are
ment transport from continental slope to deep ocean, and one documented at the Ligurian margin (Larroque et al. 2012). The
of the most common geohazards impacting on both offshore Marcel Fault shows evidence of present activity that three
infrastractures (e.g. pipeline, cables and platforms) and cos- moderate earthquakes (M3.8–M4.6) took place around this
tal areas (e.g. slope failure-induced tsunamis) (Locat and fault over the last 30 years (Larroque et al. 2012 and Fig. 1b).
Lee 2002). Slope failures are generally controlled by long- The northern upper continental slope of the Ligurian Basin
term preconditioning factors (e.g. high sedimentataion rate, is eroded by two major canyons (Var canyon and Paillon
weak layer and oversteepening) and short-term triggering canyon), which coalesce at a depth of 1,650 m (Cochonat
mechanisms (e.g. earthquake, anthropogenic activity) et al. 1993 and Fig. 1b). A single channel was formed at the
(Sultan et al. 2004). However, the exact causes for the confluence of the two canyons and divided into three parts:
different slope failure styles are still poorly understood. upper, middle and lower valleys. The walls of Upper Valley are
The Ligurian margin, northwestern Mediterreanean Sea, is highly dissected by small retrogressive failure events (Migeon
one of most intensely studied natural laboratories for landslide et al. 2011) such as that west of Cap Ferrat Ridge called
initiation in seismically active areas because of its steep topog- Western Slide (WS). It is characterized by shallow headwalls
raphy with numerous landslide scars of different size. Previous (<30 m) with high slope gradients of ~4.6 (Fig. 1c and
slope stability analyses in the region mainly focused on the 1979 Fig. 2a). A slope failure east of Cap Ferrat Ridge is termed
Nice Airport Slide or the upper slope of Ligurian margin Eastern slide (ES) and shows deep slide scars (80–120 m) and a
(Cochonat et al. 1993; Mulder et al. 1994; Sultan et al. 2004; lower slope gradient of ~3 (Fig. 1c and Fig. 2b).
Dan et al. 2007; Leynaud and Sultan 2010; Stegmann et al. Recent processes of sediment transport and deposition in
2011). This study presents two distinct slides (WS and ES) along the Var Upper Valley were mainly dominated by
the deeper slope of Ligurian margin (1,500–2,000 m below hyperpycnal-flow activity, failure-induced turbidity currents,
seaflow (mbsf)). Klaucke and Cochonat (1999) and Migeon and hemipelagite emplacement (Migeon et al. 2011). The
et al. (2011) concentrated on the mophologies of slope failure lithostratigraphic succession of WS is characterized by
and qualitatively identified their triggering mechanisms. Kopf homogenous, fine-grained hemipelagic clayey silt with some
et al. (2008) and Förster et al. (2010) charatererized the coarse-grained sand intervals (Kopf et al. 2008 and Fig. 3a).
architecture and evolution of the slope failures. Our study Areas east of Cap Ferrat Ridge are not connected to major
presents geotechnical properties of sediments from undeformed, fluvial input of the Var system and receive only hemipelagic
headwall and deposit areas of WS and ES. Those results are used sediments (Klaucke et al. 2000). The sediments are generally
for infinite slope stability of undeformed sediments under composed of well-bioturbated, homogenous, fine-grained
various conditions to (1) identify the preconditioning factors hemipelagic deposits (Kopf and Cruise 2008 and Fig. 3b).
and (2) quantify the influence of earthquakes as a key factor in
slope failing mechanisms in this densely populated area.
Materials and Methods

Geological, Geomorphological and Lithological Laboratory Tests

Setting
The principal data set for this study is based on six gravity
The Ligurian Basin is considered as a back-arc basin, that cores from undeformed slope, headwall and deposit areas of
formed by continental rifting and drifting during the late Oli- the WS and ES events. Water content was measured by a
gocene from the southeastward rollback of the Apennines- GeoTeK Multi Sensor Core Logger (MSCL) on the archive
Maghrebides subduction zone (Larroque et al. 2012 and halves at 2 cm intervals. Undrained shear strength (Su) was
references therein). Currently, active basin deformation occurs estimated using a Mennerich Geotechnik (Germany) vane
offshore at a slow rate of ~1.1 m/ka NNW-SSE, which shear apparatus and Wykeham Farrance cone penetrometer.
Geotechnical Characteristics and Slope Stability Analysis on the Deeper Slope of the. . . 551

Fig. 1 (a) Map showing the location of the study area, red circles from Larroque et al. 2012). (c) Slope-gradient map of WS and ES.
indicate earthquake records of the Ligurian margin from 1980 to 2010 Circles indicate core locations. Dashed white lines mark the headwalls
(catalogue from the Bureau Central Sismologique Français). (b) Bathy- of both slides (revised after Förster et al. 2010). Black lines indicate the
metric map of deeper slope of Ligurian margin with focal mechanisms locations of seismic profiles shown in Fig. 2b
of the moderate earthquakes associated with the Marcel Fault (taken

Grain size distribution analysis using the Beckman Coulter Slope Stability Analysis
LS 13320 particle size analyzer and Atterberg limits using
the Casagrande apparatus and rolling thread method were The 1D infinite slope stability analysis is used to calculate
carried out. Oedometer tests were performed using a GIESA the factor of safety (FS). For static conditions the FS calcu-
uniaxial incremental loading oedometer system. The drained lation after Morgenstern (1967) follows:
sediment strength parameters (cohesion c0 and internal fric-
tion angle ϕ0 ) were determined using a displacement- Su
FS ¼ ðundrainedÞ ð1Þ
controlled direct shear apparatus built by GIESA (Germany). γ 0 z sin θ cos θ
552 F. Ai et al.

c0 þ γ 0 zð cos 2 θ λÞ tan ϕ0

FS ¼ ðdrainedÞ ð2Þ
γ 0 z sin θ cos θ

Where θ is slope angle and λ* is overpressure ratio

(λ* ¼ Δu/σ0 vh), Δu is overpressure, σ0 vh is vertical effective
stress for hydrostatic conditions (σ0 vh ¼ γ0 z). γ0 is buoyant
weight, z is overburden depth.
Pseudostatic analysis was used for evaluation of slope
stability under earthquake, which is assumed the integrated
horizontal ground acceleration k g (where k is the seismic
coefficient and g is the acceleration due to gravity) to be applied
over a time period long enough for the induced shear stress to
be considered constant while the overpressure that may be
generated during an earthquake is not taken into account for
the slope stability analysis (see Mulder et al. 1994):
Fig. 2 Seismic profiles of WS (a) and ES (b) (modified after Förster
et al. 2010). Note that the bulge in panel B does not show the real Su
morphology of the headwall but is an artefact because the profile FS ¼ ðundrainedÞ ð3Þ
γ 0 z½ sin θ cos θ þ kðγ=γ 0 Þ cos 2 θŠ
crosses the flank of slope

c0 þ γ 0 zð cos 2 θ λÞ tan ϕ0

FS ¼ ðdrainedÞ ð4Þ
γ 0 z½ sin θ cos θ þ kðγ=γ 0 Þ cos 2 θŠ

Where γ is unit weight.

Results

Physical and Geotechnical Properties

Water content and undrained shear strength of sediments are

presented in Fig. 3. Sediments from undeformed slopes have
high values of water content (~60 %), while lower values
(~30 %) are seen in deeper parts of sediment cores from the
headwall. Sediments from the ES deposit area have similar
water content as sediments from the undeformed upslope
region. Low water content (~30 %) of sediments from the
deposit area of WS is attributed to coarse-grained materials.
Undrained shear strength of sediments from undeformed
slope gradually increase with depth to ~20 kPa at 5 m core
depth with value of Su/σ0 vh ranging between 0.2 and 0.4
which indicate normal consolidated state for marine
sediments (Cochonat et al. 1993) Sediments from headwall
have low shear strength (0–20 kPa) and increase rapidly up
to ~200 kPa at 3.65 m for WS and 1.o m for ES.
The dominant lithology is clayey silt (with ~20 % clay)
with an intermediate plasticity according to our Atterberg limit
measurements. Oedometer tests indicate sediment from the
undeformed slope of WS is underconsolidated (overconso-
Fig. 3 Lithology, Water content as representative for physical
properties and undrained shear strength of the sediments from WS (a) lidation ratio (OCR) ¼ σ0 pc/σ0 vh) ¼ 0.62, λ* ¼ 1—OCR
and ES (b) ¼ 0.38) and normally consolidated (OCR ¼ 0.99, λ* ¼ 0)
Geotechnical Characteristics and Slope Stability Analysis on the Deeper Slope of the. . . 553

Slope Stability Analysis

Factors of safety for four different scenarios were calculated

using Eqs. (1–4) with two parameters changing within a
certain range while all others were kept constant (for details
see Table 1). The undrained shear strength-depth relation
was obtained using fall cone tests data with linear regression.
We assume λ* of WS is 0.38 due to underconsolidated state
and no overpressure in ES because of its normal consolida-
tion state. Our data suggest that both slopes appear to be
presently stable under both undrained and drained static
conditions. The results further indicate that the slope angle
has a stronger influence on slope stability than slope failure
depth (Fig. 6). The minimum horizontal acceleration coeffi-
cient required to trigger slope failure (FS ¼ 1) was back-
calculated based on Eqs. (3) and (4). For the undrained
earthquake case, a value of k ¼ 0.08 is needed to trigger
slope failure for ES, while a lower value of k ¼ 0.06 is
needed to fail the WS slope.

Fig. 4 e-log (σ0 v) curves from oedometer tests with calculated

preconsolidation stress (σ0 pc) and overconsolidation ratio (OCR)

Discussion

Preconditioning Factors of WS and ES:

Superficial Failure vs. Deep-Seated Failure

WS is affected by superficial failures with shallow headwall

(~30 m vertical displacement) while ES shows deep-seated
failure with deeper scars (~100 m). Previous studies in the
Ligurian margin have shown that the slope angle is a
governing factor for sediment failure (e.g. Cochonat et al.
1993 and Migeon et al. 2011). High slope angles (>5 )
promote regular small-volume failure events, which pre-
vent the area to build a thick, potentially unstable sediment
Fig. 5 (a) Direct shear test protocols shown as shear stress versus
horizontal displacement. (b) Mohr-Coulomb failure planes obtained
package. Sedimentation rates are assumed to be higher
from peak shear strength values near WS on the flank of Var Upper Valley than in the ES
region due to regular sediment supply by hyperpycnal
flows (Klaucke et al. 2000). On the other hand,
for ES whereas sediments below the slip surface near the hyperpycnal flows are also involved in the gradual
headwall are strongly overconsolidated (OCR ¼ 9.6 for WS, undercutting at the base of canyon walls leading to local
OCR ¼ 72.2 for ES) (Fig. 4). The calculated thickness of oversteepening (Migeon et al. 2011). High slope angles
removed overburden material are 31 m for WS and 100 m with high sedimentation rates and effect of hyperpycnal
for ES using the equation of Silva et al. (2001). This is flows promote superficial failure in WS whereas relative
consistent with the depth estimates based on seismic profiles lower slope angle, lower sedimentation rates and without
(Fig. 2). Drained direct shear test results are presented in reworking by bottom currents promote the accumulation of
Fig. 5. Values of c0 are lower in sediments from WS a thick but more stable sediment succession in the ES
(1.8–7.7 kPa) than in ES (6.7–10.7 kPa) whereas values of ϕ0 region. The latter then serve as prerequisite and sufficient
are slightly higher in sediments from WS (30.9–33.8 ) than in material resources for deep-seated failure and larger
ES (29.5–31.9 ). volumes of slid material.
554 F. Ai et al.

Table 1 Parameters used for slope stability calculations

Parameter WS GeoB12044 ES GeoB12060
US DS UE DE US DS UE DE
Su (kPa) 1.46z 32.5 1.7z 260.0
+3.3 +5.0
z (m) 1–50 30 1–300 100
θ( ) 1–10 1–10
c0 (kPa) – 5 – 5 – 9 – 9
ϕ0 ( ) – 32 – 32 – 30 – 30
γ(kN/m2) – 17.1 – – 17.4
λ* – 0.38 – 0.38 – 0 – 0
k – 0–0.1 0–0.3 – – 0–0.1 0–0.3
γ0 (kN/m2) 7.31 7.62
g (m/s2) 9.81 9.81
FS/ 2.8/ >7/ 1/ 1/ 4.5/ >7/ 1/ 1/
k – – 0.06 0.14 – – 0.08 0.23
US Undrained Static, DS Drained Static, UE Undrained Earthquake, DE Drained Earthquake, numbers in italics indicate variable parameters

Fig. 7 PGA estimates using an empirical attenuation equation after

Bindi et al. (2011). Dashed white lines indicate the 2001 earthquake
and the distances between the slope scarps of WS and ES to the
epicenter of the 2001 earthquake. Red lines indicate the PGA needed
to currently trigger slope failure at WS and ES

to assume a drained pseudostatic model (Mulder et al.

1994). Critical pseudostatic acceleration as the average
equivalent uniform shear stress imposed by seismic shak-
Fig. 6 Undrained slope stability analyses and back-calculations of
pseudostatic horizontal acceleration ratio for WS (GeoB12044) and ing represents ~65 % of the effective seismic peak ground
ES (GeoB12060). Dashed white lines indicate current mean values of acceleration (PGA) (Strasser et al. 2011). WS is more
the parameters for static analysis and values of pseudostatic horizontal vulnerable in undrained conditions where a PGA of
acceleration required to trigger slope failure (FS ¼ 1)
0.09 g (PGA ¼ 0.06 g/0.65) is sufficient to fail the slope.
In our study, PGA has been estimated using an empirical
The Influence of Earthquake to the Slope attenuation equation after Bindi et al. (2011) (Fig. 7). Over
Stability the past 30 years, earthquakes with magnitudes 3.8–4.6
have occurred around the Marcel Fault in distances as
Superficial failures frequently occur in oversteepened, close as 10 km to WS and 5.6 km to ES (Larroque et al.
underconsolidated sediments resulting from high sedimenta- 2012 and Fig. 1b). Despite this short epicentral distance,
tion rates, while deep-seated failures probably require exter- PGA induced by the M4.6 2001 earthquake (0.03 g for WS
nal constraints such as seismic loading on the sediments to and 0.05 g for ES) is still insufficiently strong to trigger
induce slope instability. When considering acceleration- instability in either WS or ES. The attenuation relationship
induced earthquakes as a static parameter, it is reasonable indicates that moderate earthquake activity of M5.0 on the
Geotechnical Characteristics and Slope Stability Analysis on the Deeper Slope of the. . . 555

spot or stronger earthquakes (e.g. M ¼ 6.1) in epicentral nice slope (N-W Mediterranean Sea). Mar Georesour Geotechnol
distances <15 km are required to fail the WS slope. In the 11(2):131–151
Dan G, Sultan N, Savoye B (2007) The 1979 Nice harbour catastrophe
ES area, moderate earthquakes with M 5.3 on the spot or revisited: trigger mechanism inferred from geotechnical
>M6.5 earthquakes at distances <15 km are required to measurements and numerical modelling. Mar Geol 245
trigger slope failure. From Mulder et al. 1994, PGA rang- (1–4):40–64
ing from 0.095 g to 0.26 g could be expected for Förster A, Spieß V, Kopf AJ, Dennielou B (2010) Mass wasting
dynamics at the deeper slope of the Ligurian margin (Southern
earthquakes with return periods ranging from 100 to France). Submarine mass movements and their consequences
1,000 years, respectively. We propose that seismic triggers andvances in natural and technological hazard research. Springer,
may have been required for the deep-seated failure in the Dordrecht, Heidelberg, London, New York, pp 66–77
ES area, but certainly also affected the instability of super- Klaucke I, Cochonat P (1999) Analysis of past seafloor failures on the
continental slope off Nice (SE France). Geo-Mar Lett 19
ficial failure in the WS region. (4):245–253
Klaucke I, Savoye B, Cochonat P (2000) Patterns and processes of
Conclusions sediment dispersal on the continental slope off Nice, SE France.
In summary, we have demonstrated how geotechnical Mar Geol 162(2–4):405–422
Kopf A, Cruise Participants (2008) Report and preliminary results of
properties of sediments and slope stability analysis of Meteor Cruise M 73/1: LIMA-LAMO (Ligurian Margin Landslide
two distinct types of slope failure (superficial failure Measurements & Observatory). Berichte Fachbereich
and deep-seated failure) control the Ligurian Margin. Geowissenschaften, Universität Bremen, 264; 161p
Consolidation test results can be used to calculate the Larroque C, Scotti O, Ioualalen M (2012) Reappraisal of the 1887
Ligurian earthquake (western Mediterranean) from
amount of sediment removed by slope failure, which is macroseismicity, active tectonics and tsunami modelling. Geophys
consistent with depth estimates from seismic profiles. The J Int 190(1):87–104
slope angle seems to have a stronger influence on slope Leynaud D, Sultan N (2010) 3-D slope stability analysis: a probability
instability than slope failure depth below seafloor. For approach applied to the nice slope (SE France). Mar Geol 269
(3–4):89–106
better assessment the potential instability in this tectonic Locat J, Lee HJ (2002) Submarine landslides: advances and challenges.
active area, dating of different failure events is mandatory Can Geotech J 39(1):193–212
to correlate these data to real seismic events. However, Migeon S, Cattaneo A, Hassoun V, Larroque C, Corradi N, Fanucci F,
the risk assessment shows that a large-size failure only Dano A, Mercier de Lepinay B, Sage F, Gorini C (2011) Morphol-
ogy, distribution and origin of recent submarine landslides of the
requires moderate earthquake magnitudes, similar to (or Ligurian Margin (North-western Mediterranean): some insights into
even lower than) those of the 1887 historical event. Given geohazard assessment. Mar Geophys Res 32(1–2):225–243
the societal loss associated with a tsunamigenic landslide Morgenstern N (1967) Submarine slumping and the initiation of tur-
at the French Riviera, more detailed work has to be bidity currents. Mar Geotechnique 189–220
Mulder T, Tisot J-P, Cochonat P, Bourillet J-F (1994) Regional assess-
carried out in this direction. ment of mass failure events in the Baie des Anges, Mediterranean
Sea. Mar Geol 122(1–2):29–45
Acknowledgments We thank the captain and crew of the RV Meteor Silva AJ, LaRosa P, Brausse M, Baxter CD, Bryant WR (2001) Stress
for their support during the cruise M 73/1. This study is funded through states of marine sediments in plateau and basin slope areas of the
DFG-Research Center/Cluster of Excellence “The Ocean in the Earth northwestern Gulf of Mexico. Offshore Technology Conference
System” as well as the Chinese Scholarship Council. We also like to Stegmann S, Sultan N, Kopf A, Apprioual R, Pelleau P (2011) Hydro-
acknowledge the reviewers, Yasuhiro Yamada and Michael Strasser, geology and its effect on slope stability along the coastal aquifer of
for their constructive remarks. Nice, France. Mar Geol 280(1–4):168–181
Strasser M, Hilbe M, Anselmetti F (2011) Mapping basin-wide
subaquatic slope failure susceptibility as a tool to assess
regional seismic and tsunami hazards. Mar Geophys Res 32
References (1–2):331–347
Sultan N, Cochonat P, Canals M, Cattaneo A, Dennielou B, Haflidason
Bindi D, Pacor F, Luzi L, Puglia R, Massa M, Ameri G, Paolucci R H, Laberg JS, Long D, Mienert J, Trincardi F, Urgeles R, Vorren
(2011) Ground motion prediction equations derived from the Italian TO, Wilson C (2004) Triggering mechanisms of slope instability
strong motion database. Bull Earthquake Eng 9(6):1899–1920 processes and sediment failures on continental margins: a geotech-
Cochonat P, Bourillet JF, Savoye B, Dodd L (1993) Geotechnical nical approach. Mar Geol 213(1–4):291–321
characteristics and instability of submarine slope sediments, the