Landslide-risk mapping in a developing hilly area with limited information on landslide occurrence

Landslide risk was assessed at the basin scale, in terms of specific landslide risk, in a region with limited information on landslide occurrence. Specific landslide risk was mapped by combining the landslide-hazard map and vulnerability of the elements at risk. The assessment of landslide hazard was based both on the spatial-temporal probability of landslide occurrence and on landslide intensity. In this regard, a major problem to resolve was the landslide-occurrence probability. Thus, several procedures were performed, including remote-sensing-based techniques, photointerpretation, dendrochronology, and press news. On the other hand, vulnerability was studied based empirically on the inferred relationship between the hazard class and the likely damage that the landslide would cause. In this way, vulnerability was shown in terms of expected degree of loss by using property-loss criteria. The results reveal that specific landslide risk in the area is mainly low, since just 2.26% of zone showed either a possible or a likely specific landslide risk. Buildings, irrigation structures, and main roads cover 2.6% of the study zone while representing approximately 80% of the specific landslide risk. By contrast, croplands cover almost 50% of the area but represent only 7.1% of specific landslide risk. The results indicate that, although this empirical methodology is subject to some level of uncertainty, it can provide reasonable estimates of specific landslide risk over a large area despite lacking information on landslide occurrence.