Extraction of Landslide Areas in the Northern Mountainous Region of Nepal Using SAR and Optical Sensors

Abstract

In April 2015, an earthquake of Mw.7.8 occurred in Nepal. Landslides occurred frequently in the northern areas and severely affected the resident ethnic minorities. Because the affected area is widely spread, the application of remote sensed data is effective for its extraction. However, in steep mountainous areas, the observation direction and terrain shape greatly influence the extraction results of optical sensors or SAR data. Thus, this study analyzed the effectiveness of various landslide extraction methods and proposed a method that can correctly extract the areas using satellite data. As the results confirm that steep terrain shapes restricted use of SAR, and thus, land-cover classification using optical sensors is the most effective approach at present.