Automatic Detection and Classification of the Uplifted Marine Terrace by Clustering DEM Dataset
Abstract
We devised a new marine terrace classification method using the digital elevation model (DEM) and applied it to the Holocene marine terrace in the Boso Peninsula, central Japan, named Numa terrace. The Numa terrace is believed to be formed by recurrent interplate earthquakes (Kanto earthquakes). However, several recent geological surveys raised questions to this traditional interpretation and a re-investigation has been required. In this study, we proposed a new statistic method classifies the generation of paleo-shoreline angles by K-means clustering. As a result, we obtained the elevation distribution for approximately 30 km along the coastline. The revealed relative heights show comparable distributions with each other and imply their consistent deformation source. Considering the diverse formation interval times suggested by the dating survey, the traditional interplate slip model is now less accountable for the formation-history. The future task is to construct a physically consistent model explains this newly obtained geological dataset.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2019
- Bibcode:
- 2019AGUFM.T51H0388K
- Keywords:
-
- 4315 Monitoring;
- forecasting;
- prediction;
- NATURAL HAZARDS;
- 7230 Seismicity and tectonics;
- SEISMOLOGY;
- 8170 Subduction zone processes;
- TECTONOPHYSICS;
- 8488 Volcanic hazards and risks;
- VOLCANOLOGY