Earthquake-induced landslide hazard monitoring and assessment using SOM and PROMETHEE techniques: A case study at the Chiufenershan area in Central Taiwan
Abstract:Monitoring and assessment of landslide hazard is an important task for decision making and policy planning in the landslide area. Massive landslides, caused by the catastrophic Chi-Chi earthquake in 1999, occurred in Central Taiwan, especially at Chiufenershan area in Nantou county. This study proposed two useful indicators coupled with the Self-organizing map (SOM) neural network and the Preference Ranking Organization METHod for Enrichment Evaluations (PROMETHEE) technique to quickly extract accurate post-quake landslides from multi-temporal Systeme Probatoire de l'Observation de la Terre (SPOT) images. A GIS-based system was developed to simplify and integrate the procedures such as image pre-processing, the SOM training, the PROMETHEE calculation, landslide extraction and accuracy assessment. The evaluated result shows that the landslide area soon after the earthquake is 209.50 ha (Kappa coefficient 96.88%). Over seven years of vegetation recovery, the denudation area has declined to 112.64 ha (Kappa coefficient 90.64%). Most earthquake-induced landslides could be restored by natural vegetation succession. The developed system is a useful decision-making tool for landslide area planning.
Document Type: Research Article
Affiliations: Department of Environment and Disaster Management, Ming Dao University, Changhua County 52345, Taiwan
Publication date: 2008-01-01