Modeling typhoon event-induced landslides using GIS-based logistic regression: A case study of alishan forestry railway, Taiwan

  • Chen S
  • Chang C
  • Chan H
 et al. 
  • 6


    Mendeley users who have this article in their library.
  • 5


    Citations of this article.


This study develops a model for evaluating the hazard level of landslides at Alishan Forestry Railway, Taiwan, by using logistic regression with the assistance of a geographical information system (GIS). A typhoon event-induced landslide inventory, independent variables, and a triggering factor were used to build the model. The environmental factors such as bedrock lithology from the geology database; topographic aspect, terrain roughness, profile curvature, and distance to river, from the topographic database; and the vegetation index value from SPOT 4 satellite images were used as variables that influence landslide occurrence. The area under curve (AUC) of a receiver operator characteristic (ROC) curve was used to validate the model. Effects of parameters on landslide occurrence were assessed from the corresponding coefficient that appears in the logistic regression function. Thereafter, the model was applied to predict the probability of landslides for rainfall data of different return periods. Using a predicted map of probability, the study area was classified into four ranks of landslide susceptibility: low, medium, high, and very high. As a result, most high susceptibility areas are located on the western portion of the study area. Several train stations and railways are located on sites with a high susceptibility ranking. © 2013 Sheng-Chuan Chen et al.

Get free article suggestions today

Mendeley saves you time finding and organizing research

Sign up here
Already have an account ?Sign in

Find this document


  • Sheng Chuan Chen

  • Chia Chi Chang

  • Hsun Chuan Chan

  • Long Ming Huang

  • Li Ling Lin

Cite this document

Choose a citation style from the tabs below

Save time finding and organizing research with Mendeley

Sign up for free