Application of a semiquantitative and GIS-based statistical model to landslide susceptibility zonation in Kayangan Catchment, Java, Indonesia

Kayangan Catchment, one of the extremely landslide susceptible areas in Indonesia, is situated on the eastern flank of Menoreh Mountain in Yogyakarta Province on the island of Java. Landslides cause land and infrastructure damages because of their frequency in human settlements. The objectives of this study are twofold: (1) to analyze the spatial distribution of landslides and its correlation using terrain parameters; and (2) to analyze landslide susceptibility using both semiquantitative and statistical methods, i.e., analytical hierarchy process (AHP) and information value (IV) methods. Nine parameter maps were introduced to assess landslide susceptibility. The parameter maps and landslide distribution map were spatially overlaid to calculate the contribution of each parameter to landslide susceptibility. The landslide susceptibility map encompassed four different categories: very high, high, medium, and low susceptibility. The map was validated through a success rate curve by determining the area under the curve using existing landslide events. The success rate curves indicated that the IV was more accurate than the AHP, although both of them had high correlations. Both methods show that the precondition factors represented approximately 80% of the influence on landslide occurrence, with the remaining 20% attributed to the triggering factors, primarily rainfall and seismic factors.