Identification of active landslide blocks with optimized hotspot analysis using surface roughness index

Abstract

  This study introduces a method to detect active deep-seated landslides with eigenvalue ratio calculated from airborne LiDAR survey data acquired in different survey years. Applying it to the area prone to deep-seated landslides in Biratori Town, Hokkaido, where LiDAR data for 2009, 2015, and 2016 are available, field survey confirmed that the emergence of eigenvalue ratio in the range of 2.5-5.0 between the years agreed with slope activity on the spots. Optimized hotspot analysis was then applied for the spatial distribution of the cells. Hotspots suggested from the analysis successfully spotted active landslide mass.