During past two decades particularly after the 2004 Mid-Niigata Prefecture Earthquake, GIS analyses of co-seismic landslides in relation to geological and geomorphological settings have been improved, and studies on mechanical modeling of such landslides have been considerably advanced. However, these analysis using GIS alone was insufficient to complex factor analysis because of the problems in usage procedures of statistical tools and PC memory, even though GIS can support analysis of a lot of statistical layers. In addition, verification of landslide susceptibility mapping by GIS has not been inspected by a validity model up to now.
  In this paper, we have developed new analytic techniques based on buffer movement and blunder probability which clarify relevant factors of landslide occurrences by both earthquake and heavy rainfall and facilitate risk assessment supplementing GIS. Buffer movement analysis is the technique that collects GIS statistical data in the circles with radius R moving on the plan map. Blunder probability analysis is one of the objective functions to judge the compatibility of the landslide susceptibility mapping. We describe the methodology of these analysis and the results of AHP landslide susceptibility mapping as a case study of the landslides and slope failures by the 2008 Iwate-Miyagi Nairiku Earthquake.

View full abstract

  Geological and geomorphological factor analysis of landslides induced by the Mid Niigata Prefecture earthquake in 2004 and development of landslide susceptibility mapping model were carried out by using of buffer movement analysis and blunder probability analysis (Hamasaki et al., 2015). Each index value as a geological and geomorphological factor of earthquake-induced landslides was collected by buffer movement analysis from original GIS data. In addition, best combination and weight value among each index was evaluated based on blunder probability analysis. As a result, distribution of calculated susceptibility score was well accordance with distribution of actual landslide occurrence sites.

View full abstract

  Iwaki City in the southeast corner of Fukushima Prefecture in eastern Japan experienced an active-fault earthquake, named the Fukushima Hamadori earthquake (JMA magnitude 7.0), on April 11 in 2011. This earthquake was an “induced-earthquake” that occurred as a result of the Great Tohoku earthquake on March 11 in 2011. Two surface active faults appeared, and many landslides and surface slope failures were induced by the earthquake motion. We developed a landslide map to characterize the distribution of landslides using aerial and satellite images. This map shows that the landslides were relatively distributed around the hanging-wall side of the faults, which appear to be normal faults. The previous studies have shown that most of the landslides were caused by reverse fault type earthquakes, which were concentrated around the hanging-wall side and were under the influence of the “hanging-wall effect.” We think that the distribution of landslides caused by normal fault types during the time period under investigation was influenced by the “hanging-wall effect” similar to landslide distribution caused by reverse fault types.

View full abstract