 |
| Titel |
Do landslides follow landslides? |
| VerfasserIn |
Jalal Samia, Arnaud Temme, Arnold Bregt, Jakob Wallinga, Fausto Guzzetti, Francesca Ardizzone, Mauro Rossi |
| Konferenz |
EGU General Assembly 2016
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 18 (2016) |
| Datensatznummer |
250121411
|
| Publikation (Nr.) |
 EGU/EGU2016-125.pdf |
|
|
|
|
|
| Zusammenfassung |
| Landslide susceptibility maps are typically obtained by quantifying relations between landslides and conditioning attributes. Here, we take a fundamentally different starting point: path dependency and self-organization, i.e. the effect of landslides on landslides. We test two hypotheses: first, that landslides do preferentially follow landslides, and second, that follow-up landslides are different from those that do not follow other slides.
Results indicate that there is indeed a considerable amount of overlap among landslides that affect the overall affected area by landsliding. This is more than expected: the number of overlaps among landslides is more than would occur of slides were randomly placed in the study area. Overlaps of slides with previous slides occur frequently within a period of about ten years after a previous slide, yet decrease considerably over time. Also the second hypothesis is confirmed: follow-up landslides indeed have different properties in terms of power law and shape than those that are not associated. Particularly, follow-up landslides are larger and more elongated than non-follow up landslides. Moreover, after fitting an inverse gamma function to the magnitude-frequency distributions of follow-up and non-follow-up slides, it was found that the alpha parameter that controls the prevalence of very extreme events, is much larger for follow-up slides than for non-follow-up slides. Also the rollover value is substantially larger for follow-up landslides than non-follow up landslides .
The prevalence of follow-up slides in the first approximately ten years after a previous slides, and the fact that follow-up slides are different from other slides, should have implications for susceptibility studies. Apparently, susceptibility (conventionally a purely spatial concept) changes with the time since previous landslides happened. We explore possible mechanisms for this that may allow us to include these temporal changes in landslide susceptibility prediction. |
|
|
|
|
|
|