A geometry-modelling method to estimate landslide volume from source area

Eng Choon Leong*, Zhuoyuan Cheng

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

5 Citations (Scopus)

Abstract

Knowledge of landslide volume is important to understand the extent of damages and evaluating methods of remediation. However, the volume of landslide is difficult to quantify due to its scale and challenges encountered in conventional surveying. Various studies using satellite and aerial images have been conducted to empirically relate volume (i.e., displaced mass) of a landslide to its area through a power-law. However, there are many existing empirical relationships, and the volume estimate may differ substantially. In this study, firstly it is demonstrated that the empirical area-volume power-law relationships could be rationalized by a geometrical and mathematical basis. The empirical relationships in the literature are shown to be bounded by the volumes of “idealized” landslides where the slip surface is either spherical or elliptical. Secondly, a geometry-modelling method is proposed to estimate the volume of a landslide from satellite and aerial images without the need for digital elevation models. Using this method, landslide volume can be expediently estimated, and it yields better accuracy than empirical area-volume power-law relationships.

Original languageEnglish
Pages (from-to)1971-1985
Number of pages15
JournalLandslides
Volume19
Issue number8
DOIs
Publication statusPublished - Aug 2022
Externally publishedYes

Bibliographical note

Publisher Copyright:
© 2022, Springer-Verlag GmbH Germany, part of Springer Nature.

ASJC Scopus Subject Areas

  • Geotechnical Engineering and Engineering Geology

Keywords

  • Aerial images
  • Ellipsoid
  • Geometry modelling
  • Landslide
  • Power-law
  • Satellite images
  • Volume

Fingerprint

Dive into the research topics of 'A geometry-modelling method to estimate landslide volume from source area'. Together they form a unique fingerprint.

Cite this