GIS-based analytical hierarchy process, weight of evidence and logistic regression models for the landslide susceptibility predicting in Echorfa Region (northwestern of Algeria)

Authors : Roukh ZİNE EL ABİDİNE, Nadjı ABDELMANSOUR

Pages : 31-55

Doi:10.19111/bulletinofmre.1035480

View : 18 | Download : 12

Publication Date : 2023-04-17

Article Type : Research Paper

Abstract :The main objective of this study is destined to combine the Analytical Hierarchy Process insert ignore into journalissuearticles values(AHP);, Weight of Evidence insert ignore into journalissuearticles values(WOE);, Logistic Regression insert ignore into journalissuearticles values(LR); methods and geographic information system insert ignore into journalissuearticles values(GIS); to predict landslide susceptibility of the Echorfa region insert ignore into journalissuearticles values(northwestern of Algeria);. Nine factors such as slope, aspect, lithology, distance to faults, lineaments density, distance to the streams, precipitations, land use and altitude are included in landslide susceptibility evaluation process. A detailed landslide inventory map was established by satellite images and filed surveys. Three landslide susceptibility maps are established using the different statistical models. Five landslide susceptibility categories are generated by the GSI classification nil, low, moderate, high and very high susceptibility. The performance of the different models in landslide susceptibility is calculated based in the area under curve of the Receiver Operating Characteristic insert ignore into journalissuearticles values(ROC); which give a satisfactory result. The results showed that the WOE is more performance than the two other techniques. The produced landslide susceptibility maps provide important spatial information about landslide prone area, where the constructed map’s content will help the decision makers in land use planning.
Keywords : Analytical Hierarchy Process, Landslide, Logistic Regression, Susceptibility Mapping, Weight of Evidence