A GIS-based factor clustering and landslide susceptibility analysis using AHP for Gish River Basin, India

Landslide susceptibility map provides a useful tool to the decision-makers to prevent and mitigate landslide hazards. For this study 16 spatial parameters and past landslide inventory have been taken into consideration and these are categorized under six factors clusters. For providing relative importance to the parameters modified analytic hierarchy process is taken into consideration. Landslide susceptible zone (LSZ) is prepared compositing all those multiparametric spatial data layers. The obtained result shows that 7.80% area of total basin is highly susceptible for landslide. Correlation and regression analysis suggests that lithological factors cluster is the dominant one for determining very high LSZ. The validation shows that very high LSZ is associated with very high landslide frequency density. Besides this, receiver operating characteristics curve also shows 90.20% predicted area under the curve. So, this model can be treated as valid.