Share Email Print

Proceedings Paper

Landslide susceptibility mapping at Dodangeh watershed, Iran using LR and ANN models in GIS
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Landslide is among the most common geologic hazard around the world. They cause many formidable damages and numerous deaths annually. There are multiple causes for every landslide. Landslide causes are classified into three categories, namely, physical, natural, and human. Maps are a valuable and appropriate tool for presenting information on landslide. Landslide susceptibility maps is one of the most useful source of information for landuse planners. A landslide susceptibility map represents zones that have the potential for landslide. These zones are identified by correlation between the past distribution of landslide occurrence and conditioning factors that contribute to landslide. This study employs logistic regression (LR) and artificial neural networks (ANN) models to assess landslide susceptibility in Dodangeh Watershed, Mazandaran Province, Iran. The spatial database included landslide inventory, altitude, slope angle and aspect, plan and profile curvatures, distance from faults and from stream, stream power index (SPI), topographic wetness index (TWI), sediment transport index (STI), terrain roughness index (TRI), landuse and lithology. Validation of the models using receiver operating characteristics and overall accuracy indicates that both models display satisfactory performance, and LR model exhibits the most stable and best performance. Given the outcomes of the study, the LR model, which has an AUC value of 0.872 and an overall accuracy of 82.59%, and the ANN model, which has a AUC value of 0.77 and an overall accuracy of 71% are promising techniques for landslide susceptibility mapping.

Paper Details

Date Published: 9 October 2018
PDF: 10 pages
Proc. SPIE 10790, Earth Resources and Environmental Remote Sensing/GIS Applications IX, 107901D (9 October 2018); doi: 10.1117/12.2501473
Show Author Affiliations
Bahareh Kalantar, RIKEN (Japan)
Naonori Ueda, RIKEN (Japan)
Husam Abdulrasool H. Al-Najjar, Univ. Putra Malaysia (Malaysia)
Mohammed Oludare Idrees, Univ. Putra Malaysia (Malaysia)
Alireza Motevalli, Tarbiat Modares Univ. (Iran, Islamic Republic of)
Biswajeet Pradhan , Univ. of Technology, Sydney (Australia)

Published in SPIE Proceedings Vol. 10790:
Earth Resources and Environmental Remote Sensing/GIS Applications IX
Ulrich Michel; Karsten Schulz, Editor(s)

© SPIE. Terms of Use
Back to Top