Abstract

Sediment disasters caused by deep-seated catastrophic landslides have become a serious problem in Japan. Thus, it is necessary to identify the slopes with higher risk to deep-seated catastrophic landslides in order to establish disaster prevention planning. In this study, a method based on both self-organizing map and Hayashi's second method of quantification is proposed. Slopes with higher risk to deep-seated catastrophic landslides are then identified by applying the proposed method to 142 slopes in Nara Prefecture. Furthermore, the slopes are prioritized according to the sample scores of Hayashi's second method of quantification.

Keywords:
concave, undercut slope, risk assessment, topographical characteristic, knick line, machine learning, categorical data, river, topographical map, deep-seated catastrophic landslide
Subjects:
Information Management and Systems, Artificial intelligence
This content is only available via PDF.
2014. International Society of Offshore and Polar Engineers
You can access this article if you purchase or spend a download.