Volume 8, Issue 15 (9-2017)                   jwmr 2017, 8(15): 161-170 | Back to browse issues page


XML Persian Abstract Print


Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

(2017). Prioritization of Landslide-Conditioning Factors and its Landslide Susceptibility Mapping using Random Forest New Algorithm (Case Study: A Part of Golestan province) . jwmr. 8(15), 161-170. doi:10.29252/jwmr.8.15.161
URL: http://jwmr.sanru.ac.ir/article-1-852-en.html
Abstract:   (3726 Views)
Landslide as a natural hazard is very dangerous especially in mountainous areas that result in loss of human life and property around the world. Iran is always exposed to landslide hazard especially in the north and west because of climatic and topographic conditions. The aim of this research is prioritization of landslide-conditioning factors and its landslide susceptibility mapping in the part of Golestan Province using random forest new algorithm. At first, landslide locations were identified using field survey, historical report, and Google earth. In total, 78 landslide locations were identified and divided into two parts for modeling (70%) and validation (30%). Eleven factors of landslide-conditioning including slope aspect, altitude, distance from streams, distance from faults, distance from roads, lithology, land use, slope-length, plan curvature, precipitation, and slope angle were prepared. The relationships between the effective factors and the landslide inventory map were calculated using the random forest algorithm, and then landslide susceptibility map was prepared in the GIS environment. Prioritization of landslide-conditioning factors showed that distance from road, distance from faults, and altitude have the most effect on landslide occurrence respectively. Finally landslide susceptibility map produced by random forest model were divided to four susceptibility classes such as low (29.18%), moderate (33.44%), high (24.82%), and very high (12.55%). ROC curve and the area under the curvewere used for accuracy assessment of the prepared map using about 30% of landslides. Results showed that the random forest model produced reasonable accuracy in landslide susceptibility mapping with area under curve of 0.706.
Full-Text [PDF 620 kb]   (1959 Downloads)    
Type of Study: Research | Subject: Special
Received: 2017/09/19 | Accepted: 2017/09/19 | Published: 2017/09/19

Add your comments about this article : Your username or Email:
CAPTCHA

Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

© 2024 CC BY-NC 4.0 | Journal of Watershed Management Research

Designed & Developed by : Yektaweb