2265
A new approach to assess the potential for flow slide in sensitive clays
Une nouvelle approche pour évaluer le potentiel
d’écoulement d
es argiles sensibles
Thakur V., Degago S.A., Oset F., Dolva B.K., Aabøe R.
Geotechnical section, Norwegian Public Roads Administration
ABSTRACT: Sensitive clays could constitute a major threat to nearby infrastructure due to potential flow slides. A key question is if
all sensitive clays have the same potential to retrogress to the same extent and the significance of remoulded shear strength (
c
ur
) in
flow slides. This paper proposes a new approach to asses the potential for flow slides and in doing so presents a new laboratory test
procedure referred to as the quickness test. The test focuses on remoulded behaviour of sensitive clays in terms of a numerical value
referred as quickness (
Q
). Sensitive clay samples were collected from three different landslide locations. The quickness test was used
to demonstrate why sensitive clays with
c
ur
>1 kPa are not susceptible to flow slides. Based on this study, it is possible to state that a
Q
< 15 % or
c
ur
> 1 kPa seems to be the threshold limit above which the extent of the retrogression of a landslide is limited to the
initial slide. This criteria has been supported using data available from several Norwegian landslides in sensitive clays.
RÉSUMÉ : Les argiles sensibles pourraient constituer une grave menace pour les infrastructures en raison de coulées potentielles.
Une question clé est de savoir si toutes les argiles sensibles ont un potentiel de rétrogression équivalent et
de d’estimer l’amplitude de
la résistance de cisaillement remaniée (
c
ur
) des les coulées. Cet article propose une nouvelle approche pour évaluer le potentiel de
coulées et présente ainsi une nouvelle procédure de laboratoire appelée le test de rapidité. Le test se concentre sur le comportement de
l’argile sensible remaniée en termes de valeur numérique dénommé
e rapidité (
Q
). Des échantillons d'argiles sensibles ont été
recueillis à partir de trois sites différents où il y a eu des glissements de terrain. Le test de rapidité a été utilisé pour démontrer
pourquoi les argiles sensibles avec
c
ur
> 1 kPa ne sont pas sujettes à des coulées. Sur la base de cette étude, il est possible d'affirmer
que
Q
<15% ou
c
ur
> 1 kPa semble être le seuil au-delà duquel l'ampleur de la rétrogression d'un glissement de terrain est limitée à la
rupture initiale. Ce critère se base sur des données disponibles à partir de plusieurs glissements de terrain dans les argiles sensibles
norvégiens.
KEYWORDS: Flow slide, sensitive clays, quickness, remolded shear strength, landslide.
1 INTRODUCTION
Soft sensitive clays are normally associated with loss of
stability as well as substantial ground deformation, which can
lead to structural damage and jeopardize the overall stability of
an area. It is worth mentioning the statement by Bhishop in
1973 that “
Rotational slips alone may cause damage to services
and property, but seldom involve loss of life. It is the subsequent
development into a flowslide which has been responsbile in
each case where disaster has occured”
. In the sensitive clay
deposits of Scandinavia and eastern Canada, landslides are
particularly destructive, due to the possibility of small
landslides initiating a flow slide, which may involve massive
soil movements in the order of millions of cubic meters.
For flow slides to occur after an initial slide, it is important
that at least the following two criteria are fulfilled (Tavenas et
al. 1983):
The slide debris should be sufficiently remoulded.
The slide debris should be able to flow out of the slide
area if remoulded.
There may be additional factors, such as the topography and
the stability of the area behind the initial slide zone. However, if
the two criteria mentioned above are not fulfilled, then vast
landslides, such as those listed in Table 1, are unlikely to occur.
An overview of Norwegian landslides as shown on Table 1
clearly indicates that soft sensitive clays must be handled
carefully.
However, the term ‘sensitive clay’ encompasses a
wide range of clays depicting significant variations in
engineering behaviour such as the remoulded shear strength
(
c
ur
). Therefore, a key question is whether all sensitive clays
have the same potential to create flow slide and the significance
of
c
ur
for occurrence of flow slide? Answer to this question has
a direct relavance in the selection of an appropriate safety level
for constructions, measures and in hazard mapping.
A relationship between the
c
ur
and the soil sensitivity (
S
t
) as
well as between
c
ur
and the liquidity index (
I
L
) has been used to
evaluate the potential for flow slides in sensitive clays. Mitchell
and Markell (1974) suggest a direct relationship between
c
ur
,
S
t
and the retrogression distance (
L
). They suggest that landslides
with an
L
> 100 m are observed for
c
ur
< 1 kPa. Retrogression
was zero for c
ur
> 1 kPa. Leroueil et al. (1983) and Locat and
Demers (1988) presents correlations between c
ur
and
I
L
. The
correlations are
c
ur
= (
I
L
-0.21)
-2
and
c
ur
= 1.46
I
L
-2.44
respectively. These showed that senstive clays with I
L
> 1.2 are
susceptible to flow slides. This finding is also supported by the
landslide data presented in Table 1. It must be noted that,
according to the correlations
I
L
> 1.2 is only possible when
c
ur
is
less than 1 kPa. In other words, the findings by Mitchell and
Markell (1974), Leroueil et al. (1983), Tavenas et al. (1983) and
Locat and Demers (1988) are in line with each other. Based on
the landslide data, Lebuis et al (1983) also suggested that
c
ur
< 1
kPa may define the threshold limit for occurrence of flow slides.
I
L
and
c
ur
based criteria have been widely adopted to study
the flow slide potential of sensitive clays, However, the
measurement of
I
L
demands determination of three parameters a
priori, i.e. liquid limit (
wl
), plastic limit (
wp
) and natural water
content (
w
). Notably the conventional thread-rolling method of
