 |
| Titel |
Shallow landslide preliminary risk analysis based on a rainfall event in Switzerland |
| VerfasserIn |
Pierrick Nicolet, Olivier Caspar, Marc Choffet, Marc-Henri Derron, Loris Foresti, Michel Jaboyedoff |
| Konferenz |
EGU General Assembly 2013
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 15 (2013) |
| Datensatznummer |
250082572
|
|
|
|
|
|
| Zusammenfassung |
| Central Switzerland was affected in 2005 by extreme rainfall, causing floods, debris-flows
and a large number of landslides. Indeed, more than 5’700 landslides and earth flows were
reported and produced damage reaching CHF 16 million to private infrastructure. 6 persons
were killed, among which 2 were killed by a landslide-generated debris-flow.
Comparing landslide distribution and precipitation maps, we can observe that
landslide density is highly correlated with precipitation amounts over the entire
precipitation event and with lithological units. Thus, a probabilistic model can be
deduced from the precipitation map to obtain the probability of overpassing a certain
number of landslides in each model cell (1km x 1km) for a specific precipitation
event.
To find if a landslide will produce some damage, a concept of intersection probability is
used. We assume that the landslide can occur anywhere within a polygon given by an
indicative hazard map with the same probability. Given this first assumption, if a landslide
occurs, the probability that it hits a building is given by the ratio between the surface covered
by buildings – extended by a buffer depending on the landslide dimensions – and the total
surface.
Thus, a mean annual risk estimation can be modeled from the precipitation return periods
inferred from station data. |
|
|
|
|
|
|