1507
Impact of blast vibrations on the release of quick clay slides
Impact des vibrations dues aux explosions sur les glissements de terrain dans les argiles sensibles
Johansson J., Løvholt F., Andersen K.H., Madshus C.
Norwegian Geotechnical Institute
Aabøe R.
Norwegian Public Roads Administration
ABSTRACT: Blast induced vibrations are suspected to be one of the triggering factors of quick clay slides. To better understand how
such vibrations affect sensitive clays we have analyzed vibrations from blasts near quick clay deposits and performed numerical
simulations. In addition, based on a set of cyclic laboratory tests on quick clay with anisotropic consolidation, a set of cyclic loading
contour diagrams have been established. The diagrams are used to estimate the effect of cyclic loading on the clay from blasting in
combination with the permanent stress in quick clay slopes. The above results were combined in a simplified method of estimating
vibration velocities that could cause a local failure in a slope. The velocities are in the range of 57-110 mm/s depending on degree of
strength mobilization in the slope and number of load cycles imposed by the blast vibration. Taking into account uncertainties, spatial
variability and desired safety margins we recommend a vibration limit of 25 mm/s. We further recommend that vibration monitoring
during blast operations should be performed at two separate locations, near the rock-clay boundary, with the highest vibration
amplitude in any of the vertical and the two orthogonal horizontal directions should be lower than the vibration limit.
RÉSUMÉ :
Les explosions génèrent des vibrations qui sont soupçonnées d’être un des facteurs déclencheurs des glissements de
terrain dans les argiles sensibles. Pour mieux comprendre comment de telles vibrations affectent ces argiles, nous devons analyser les
vibrations d’explosions proches des dépôts d’argile sensible et effectuer des simulations numériques. De plus, basé sur un ensemble
de tests cycliques faits en laboratoire sur les argiles sensibles avec une consolidation anisotropique, un ensemble de diagrammes de
contours de chargements cyclique a été établi. Ces diagrammes sont utilisés pour estimer les effets de chargement cyclique sur l’argile
suite aux explosions et soumises aux stress permanents des argiles sensibles sur les pentes. Les résultats ci-dessus sont combinés dans
une méthode simplifiée d’estimations des vitesses de vibrations qui peuvent causer une rupture locale de la pente. Les vitesses sont
comprises entre 57 et 110 mm/s, dépendant du degré de la force de mobilisation dans la pente et de la charge de cycles imposé par les
vibrations suite à l’explosion. En tenant compte des imprécisions, de la variabilité spatiale et des marges de sécurité désirée, nous
recommandons une limite de vibration de 25 mm/s. Nous recommandons également de surveiller les vibrations durant les explosions
à deux
endroits différents prés de l’interface roche
-argile avec enregistrement dans la direction verticale et les deux orthogonales
pour
s’assurer que les amplitudes des vibrations les plus importantes soient enregistrées
KEYWORDS: Quick clay, blast, vibration limit, cyclic loading, creep failure, local failure, global failure.
1 INTRODUCTION
In 2009, a large quick clay slide involving up to 500,000 m
3
soil, and affecting some 10 houses, was triggered by rock
blasting for a road cutting in Kattmarka near the city of Namsos
in Norway. For this event it was concluded the trigger was not
the vibrations themselves, but a large block of rock that was
punched in to the soil by the blast effect, due to unfavourable
fault planes in the rock. The landslide did however reinitiate
discussions on slope stability with respect to blast induced
vibrations in the vicinity of sensitive soils. A brief literature
survey showed that there are other events where vibrations may
have been one of the triggering factors. Yet, prior to 2009, no
regulations existed, and the understanding of blast induced
landslide events was limited.
After the 2009 Kattmarka quick clay slide, a vibration limit
of 25 mm/s was introduced in the Norwegian Public Roads
Administration’s
Handbooks (NPRA, 2011) based on
engineering considerations of soil dynamics and general cyclic
soil response. However, experimental data on cyclic behaviour
of quick clay were sparse.
To improve the understanding of the effect of blast induced
vibrations on the possible triggering of slides in sensitive clays,
a research project was initiated at NGI, with financial support
from NPRA, the Norwegian Gove
rnment’s agency for railway
services, and Norwegian Research Council. The goal of the
project is to establish vibration limit values and vibration
monitoring procedures to avoid landslides due to blast
operations, without imposing more restrictions than necessary.
The research activity has involved a brief literature study of
previous case histories, a set of laboratory tests and re-
evaluation of relevant previous laboratory tests. We have also
analyzed vibration measurements and performed numerical
analyses. Finally we have combined the above in a simplified
method of estimating vibration velocities that could cause a
local failure in a slope.
2 CASE HISTORIES
There seem to have been a common opinion in the Norwegian
geotechnical community that vibrations from rock blasting do
not have a high probability of trigging slides, even near quick
clay slopes. This has however been questioned in connection
with the Norwegian slides in Finneidfjord (NGI, 1997) and
Kattmarka (NTNU, 2009). In 1973 a large slide in Fröland,
Uddevalla in Sweden was triggered by liquefaction of thin sand
and silt layers due to vibration from blasting at an adjacent
stone quarry (Bjurström and Broms, 1973). Similar layers are
also found in Finneidfjord and Kattmarka, and are considered
common in most Norwegian quick clay deposits.
Very recently, in 2011, a slide occurred in a small ravine
with quick clay more than 100 meters from ongoing road works
near Lödöse in Sweden, involving rock blasting. The last round
had been fired less than 24 hours before (Ekström, 2012). The
peak vibration velocity has been estimated (frequency
unknown) to some 30 mm/s at the upper edge of the slide with
help of post slide measured blast vibrations.
The literature review further indicates that if blast vibrations
are assumed to have triggered a slide it is often in combination
