Abstract
Abstract: In this paper, a GIS-based methodology has been used to produce a landslide susceptibility map for a landslide-prone area located at the central northern part of Peninsular Malaysia along the E-W highway (Gerik - Jeli). The susceptibility assessment was based on a bivariate statistical method namely the statistical index by which the spatial relationship between landslides and causative factors have been evaluated. Landslide locations were identified from interpretation of aerial photographs, field surveys and previous studies conducted in the area. Spatial database for causative factors was constructed from topographical maps, geological maps, satellite data, hydrological data, soil data and field data. Ten thematic maps for factors considered in the analysis including: Slope gradient, slope aspect, elevation, distance from road, drainage density, lithology, strata dip map, foliation dip map, lineament density, and soil map were extracted from the constructed spatial database. The validation of assessment showed a satisfactory agreement between the susceptibility map and the landslide locations. The area under curve (AUC) of success rate for the model was (85.13%) and the AUC of prediction rate was (82.21%), indicating a high prediction accuracy. The influencing factors on the landslide susceptibility were evaluated qualitatively by excluding each factor from the analysis to select the positive factors that improve the prediction accuracy of the landslide susceptibility map. Factor effects revealed that all factors have relatively positive influences on the resulting landslide susceptibility map but the most effective factors on analysis are: distance from road (AUC = 0.725), lineament density (AUC = 0.756), slope gradient (AUC = 0.787), and lithology (AUC = 0.793). The statistical index (Wi) values of the considered variables pointed to the importance of three classes as the most contributing factors to landslides in the study area which are: less than 300 m distance from road (Wi >2.25) , high and very high density lineament (Wi >1.44), slope gradient 35°- 45° (Wi =0.71), and lithological unit composes of phyllite and slate (Wi = 0.68).