Proceedings Article | 12 December 2006
KEYWORDS: Landslide (networking), Roads, Lawrencium, Geographic information systems, Agriculture, Raster graphics, Remote sensing, Earthquakes, Fuzzy logic, Natural disasters
It is well known that natural disasters such as earthquakes, landslides, floods, etc. cause enormous damage to lives and
property. The assessment of risk as a potential for adverse consequences, loss, harm to human population due to the
occurrence of natural disasters, particularly the landslides in Himalayan region therefore becomes imperative. Landslide
risk assessment (LRA) techniques can be applied at different stages in the decision-making process, starting from
developmental planning on a regional scale to a particular site evaluation at local scale. The LRA depends on the
probability of landslide hazard and the vulnerability of risk elements. The landslide probability depends on both the
preparatory (i.e., inherent ground characteristics) and triggering (i.e., earthquake and rainfall) factors. Vulnerability may
be defined as the level of potential damage, or degree of loss, of risk elements subjected to landslide occurrences. The
assessment of vulnerability is somewhat subjective and on a regional scale it is largely based on the importance of risk
elements in human society. Hence, the appropriate vulnerability factor may be assessed systematically by expert
judgment. In the present study, a linguistic rule based fuzzy approach is developed and implemented to prepare the
landslide risk assessment (LRA) of Darjeeling Himalayas. The LRA has been considered as a function of landslide
potential (LP) and resource damage potential (RDP), which have been characterized by the landslide susceptibility
zonation (LSZ) map and the resource map (i.e., land use land cover map including the road network) of the area
respectively. Fuzzy membership values representing LP and RDP have been assigned to different categories of LSZ and
resource maps based on the criteria developed on a linguistic scale. Landslide risk assessment matrix (LRAM) has been
generated as a function of the fuzzy membership values, which reflects the relative risk values for different combinations
of landslide potential and resource damage potential. These landslide risk values have been classified into six different
zones namely, no risk, very low risk (VLR), low risk (LR), moderate risk (MR), high risk (HR) and very high risk
(VHR) to ultimately prepare LRA map of the area. Based on this map, a risk management action plan may be suggested
to avoid the possible risk to the resources available in the area.