2237
Landslides induced by the interaction of an earthquake and subsequent rainfall.
A
spatial and temporal model.
Glissements de terrain induits par l’interaction d’un tremblement de terre suivi de précipitations – Un
modèle spatio-temporel.
Quan Luna B., Vangelsten B.V., Liu Z.Q., Eidsvig U., Nadim F.
Norwegian Geotechnical Institute (NGI) / International Centre for Geohazards (ICG), Norway
ABSTRACT: The consideration of multiple hazards and their interaction to achieve risk reduction is a necessity since many regions are
prone to different types of threats. However, this is neither simple and straightforward nor commonly undertaken at present since dif-
ferent natural hazards are usually analyzed individually and managed separately. A common example of this is the alteration of the
density and locations of rainfall-induced landslides after an earthquake due to the extensive disturbance of surface strata. That implies
an influence of the earthquake on the soil structure that alters the disposition towards landslides. Taking this into consideration, a
model was developed that could give a rough spatial and temporal prediction of expected change in landslide hazard in an area follow-
ing an earthquake. The model is able to describe the reduced impact of earthquakes with distance from epicentre as well as how the soil
gradually regains its strength with time. These reductions are then applied to an equilibrium stability analysis in order to compute new
Factors of Safety. Although analysis schemes can be proposed and software tools can be provided to facilitate many steps, a well-
conceived and reflective approach to multi-hazard settings is still essential.
RÉSUMÉ : Etant donné que différentes régions sont sujettes à différent types de menace, l’étude de multiples facteurs de risque et de
leur interaction est nécessaire afin de réduire le risque. Ce genre d’études est néanmoins rarement entrepris de nos jours puisque chaque
risque naturel est usuellement analysé et traité séparément. Un exemple courant est l’évolution de la densité et de la localisation des
glissements de terrain induits par des précipitations suite à un tremblement de terre en raison de l’importante perturbation des couches
superficielles. Cela implique que l’influence des tremblements de terre sur la structure du sous-sol modifie la prédisposition aux glis-
sements de terrain. Compte de tenu de cela, un modèle de prédiction spatio-temporelle de la variation du risque de glissement de terrain
dans une région donnée suite à un tremblement de terre a été développé. Le modèle est capable de décrire la diminution de l’impact des
tremblements de terre en fonction de la distance à l’épicentre ainsi que la manière dont le sous-sol retrouve sa résistance initiale. Ces
diminutions sont ensuite appliquées à une analyse de stabilité de façon à calculer de nouveaux facteurs de sécurité. Bien que des pro-
grammes d’analyses puissent être proposés et que des logiciels peuvent être fournis afin de faciliter de nombreuses étapes, une appro-
che réfléchie et bien conçue des environnements à dangers multiples reste indispensable.
KEYWORDS: Landslide, Earthquake, Rainfall, Slope stability, Factor of Safety
1 INTRODUCTION
A multi-hazard analysis refers to the implementation of meth-
odologies and approaches aimed at assessing and mapping the
potential occurrence of different types of natural hazards in a
given area. Analytical methods and mapping have to take into
account the characteristics of the single hazardous events as
well as their mutual interactions and interrelations (Delmonaco
et al. 2006). The existence of relations between natural hazards
and the potentially resulting consequences is an issue of increas-
ing importance in multi-hazard studies (Kappes et al. 2010).
The specific methods to deal with related hazards are as diverse
as the terms and the phenomena falling into this category. How-
ever, a general approach can be applied where the investigation
of the individual chain of one hazard triggering the next is per-
formed (so-called cascading events).
Evidently, performing multi-hazard risk analysis is not a
simple operation. Apart from the data requirements and time-
consuming conduction of single-hazard risk studies that require
know-how from different disciplines, many further aspects have
to be considered. One important source of difficulty is the con-
trast in hazard characteristics. Hazards differ with respect to
their properties such as time of onset, duration, physical proper-
ties and extent. As a result, the modelling approaches adjusted
to the hazard specifics also contrast strongly (Bovolo et al
2009).
2 MODEL DESCRIPTION
The proposed model aims to compute and estimate the associ-
ated alteration in the landslide hazard following an earthquake
event. The model describes the reduced impact of earthquakes
with distance from epicentre as well as how the soil gradually
regains its strength with time. This was achieved by analyzing
the collected data on how seismic activity influences the critical
rainfall intensity and duration needed to trigger landslides in the
area impacted by the earthquake. Based on the gathered infor-
mation the modelling objective was met by constructing three
sub-models: A) Use empirical data to construct a sub-model
linking rainfall threshold reduction as function of time after
earthquake impact; B) Use empirical data to construct a sub-
model linking rainfall threshold reduction as function of dis-
tance from epicentre and earthquake magnitude, C) Use an ana-
lytical hydrological model to link rainfall threshold reduction to
changes in soil shear strength.
Sub-models A and B together form an empirical model for
rainfall threshold reduction as a function of magnitude, distance
and time after the earthquake. The model is based on the follow-
ing assumptions:
The effect of earthquake on the rainfall threshold reduction
decays with distance from and time after the earthquake.
The maximum rainfall threshold reduction (i.e. rainfall
threshold reduction immediately after and at epicenter) is
assumed to be a function of earthquake magnitude only.
li
i
t i t r ti
f rt
t r i f ll.
A spatial and temporal model
liss
ts t rr i i
its r l’i t r cti
’
tr
l
t t rr s ivi r ci it ti
s.
Un modèle spatio-temporel
uan Luna ., angelsten . ., Liu . ., idsvig ., adi .
or egian eotechnical Institute ( I) / International entre for eohazards (I
), or ay
