| The recent technological development of LiDAR has greatly improved the resolution with which topographic detail can be derived from DEMs. Being able to quantify fine topographic features with filters, LiDAR-derived DEMs can be used to estimate landslide conditions and prioritize slopes for remediation. To utilize LiDAR data for this purpose, it is first necessary to establish the relationships between filter values and actual landslide conditions. These relationships will vary between sites, depending on the stage of landslide development, which is in turn is influenced by the underlying bedrock type. In this study, two filters, slope angle and surface roughness (eigenvalue ratio: ER), were used to analyze the features of 8 landslides, at various stages of evolution. Slope angle is an indicator of landslide maturity, and ER represents their activity. The landslides analyzed are located in 3 different geological settings at Kii Peninsula (Kii Site) and Fukushima Prefecture (Fukushima Site). The Fukushima site consists of two sub-sites, harder (HF site) and softer (SF site) bedrock sites. The grid size of DEMs was 1 m for the Kii site and 2 m for both the HF and SF sites. Results show that the range of ER at each site expressed landslide activity. Immature (steep) and active landslides in the Kii site contained a higher proportion of low ER range (<4), in which cracked outcrops, scarps, cracks, and colluvial deposits were commonly observed in the field. On the other hand, in the HF site, the density of low ER (<3) was also high in the active slide because of cliffs and cracks. In the block, a high proportion of cells in the range of ER 4-5 expressed a variety of topographic features. Finally, in the SF site, where bedrocks are considered to be the softest among the study sites, a mature (very gentle), but still active landslide, had also formed various topographic features with a high density of ER values in the range of 4-5. At study sites, it was common that landslides evolved and became gentle as dissected areas were planed, non-dissected areas were dissected with time: the process and activity of landslides were well represented by the slope angle and ER filters. Because there is no major difference in the evolutionary process of landslides between differing lithological settings, except for progress rates and final features attained, these filters have potential to assist in estimating landslide conditions elsewhere. |
