Open AccessLandslide susceptibility mapping along road corridors in west Sulawesi using GIS-AHP models
Author(s) -
Ardy Arsyad,
Wan Abdul Hamid
Publication year - 2020
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.179
H-Index - 26
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/419/1/012080
Subject(s) - landslide , analytic hierarchy process , spatial database , cartography , raster data , field survey , raster graphics , geography , geology , spatial analysis , remote sensing , computer science , geotechnical engineering , engineering , artificial intelligence , operations research
Landslide susceptibility mapping (LSM) along road corridors is a fundamental tool for road planners and transportation decision-makers in determining the severity of slope failure areas, and in planning effective landslide prevention strategies. The accuracy of LSM relies on sufficient spatial database from the data of relevant landslide contributing factors. However, for roads located in remote areas in the Eastern Part of Indonesia, the database remains poor, leading to difficulty in mapping landslide prone areas. This study presents a framework of LSM along road corridors in the area with a limited database. The framework comprises identification and classification of several landslide contributing factors including slope angle, the distance of slope to the road, the distance of drainage to the road, the distance of faults to the road, lithology where road located, and precipitation on the area. These factors are weighted based on causative relation to landslide occurrence by using the analytical hierarchy process (AHP), then converted into the grid in a raster map by using Arc-GIS. The maps are then overlaid to generate landslide susceptibility ratings over the area. The LSM framework is tested in a landslide-prone area along road corridors of Polewali – Malabo, West Sulawesi Indonesia. The accuracy of the obtained LSM is evaluated with R-index, which is the index of how closed predicted slide area in the LSM and real slide area in the field obtained from landslide inventory undertaken during the field survey. It is found that R-index of the LSM is 91.3%, indicating a better prediction of landslide hazard with the LSM framework. The framework of LSM for road corridors would be useful for landslide hazard mitigation in the road infrastructure of developing countries.