Comparative Evaluation of GIS Based Landslide Hazard Zonation Maps Using Different Approaches

Natural hazards like landslides in lesser Himalayan region of district Kullu (HP) India, owe to its typical geomorphic setting, variations in relief, precipitation during monsoon, thick forest cover, presence of glacier and glacial lakes along the higher reaches and various anthropogenic activities. Landslide locations, types and factor influencing the slope details are collected through the field visits and past record of landslides were used for development of inventory. Ancillary data are used for generation of different thematic layers (aspect, drainage density, geology, land cover, lineament density, relief, slope, and soil). Weight of evidence (WOE), information value(IV) and frequency ratio (FR) methods are used for generation of landslide hazard zonation (LHZ) maps which were classified into four zones namely high, moderate, low and very low. Validation and comparison of these hazard maps was done through the successive rate curve method which showed almost similar results, this proved the accuracy of these methods for generation of LHZ maps (WOE 80.01%, IV81.54% and frequency ratio 83.21%).