Comparison of Single Event-based and Multiple Event-based Landslide Modeling Method using Artificial Neural Network
Abstract
Landslide hazard assessment is able to be conducted by analyzing historical landslide events which can be caused by different triggers, including heavy rainfall, earthquake and human activities. For landslide modeling, many studies have focused on developing landslide models based on single landslide event (rainfall or earthquake). However, considering the variations of different events, regarding their magnitude and distribution of induced landslides, this study examines the difference of modeling results between two approaches: (1) single event-based modeling (SEB): developing a landslide model using data from one typhoon event, and validate the model with other two typhoon events, and (2) multiple event-based modeling (MEB): developing a landslide model using data from two events and validate the model with the other event. Yushan National Park (YNP), located in central Taiwan, was chosen as an ideal test site where has been continuously affected by typhoons with different magnitude, creating a complex landslide environment. In this research rainfall data of three typhoons that struck YNP in 1996, 2001 and 2009 were collected to develop the SEB and MEB landslide models using Artificial Neural Network (ANN). The model outcomes were validated by manually interpreted landslide inventories of each typhoon event, and the result shows multiple event-based models perform better than single ones. For SEB, the validation accuracies vary form 68% - 88%, and 87% - 91% for MEB. In average, the validation set accuracies are: 78% for SEB, and 89% for MEB. This study demonstrates the importance of considering multi-events in landslide modeling, and also reveals the applicability of ANN method in landslide hazard assessment.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFMNH21B0817C
- Keywords:
-
- 1810 Debris flow and landslides;
- HYDROLOGYDE: 4302 Geological;
- NATURAL HAZARDSDE: 4303 Hydrological;
- NATURAL HAZARDSDE: 4315 Monitoring;
- forecasting;
- prediction;
- NATURAL HAZARDS