HR: 1340h
AN: H43E-1665    [Abstracts]
TI: Analysis of the effects of geological and geomorphological factors on earthquake triggered landslides using artificial neural networks (ANN)
AU: * Kawabata, D
EM: d-kawabata@aist.go.jp
AF: Geological Survey of Japan, AIST, National Institute of Advanced Industrial Science and Tecnology (AIST) AIST Tsukuba Central 7 1-1, Higashi 1-Chome, Tsukuba, 3058567, Japan
AU: Bandibas, J
EM: joel.bandibas@aist.go.jp
AF: Geological Survey of Japan, AIST, National Institute of Advanced Industrial Science and Tecnology (AIST) AIST Tsukuba Central 7 1-1, Higashi 1-Chome, Tsukuba, 3058567, Japan
AB: The occurrence of landslide is the result of the interaction of complex and diverse environmental factors. The geomorphic and geologic features, rock types and vegetative cover are important base factors of landslide occurrence. However, determining the relationship between these factors and landslide occurrence is very difficult using conventional mathematical analysis. The use of an advanced computing technique for this kind of analysis is very important. Artificial neural network (ANN) has recently been included in the list of analytical tools for a wide range of applications in the natural sciences research fields. One of the advantages of using ANN for pattern recognition is that it can handle data at any measurement scale ranging from nominal, ordinal to linear and ratio, and any form of data distribution (Wang et al., 1995). In addition, it can easily handle qualitative variables making it widely used in integrated analysis of spatial data from multiple sources for predicting and classification. This study focuses on the definition of the relationship between geological factors and landslide occurrence using artificial neural networks. The study also focuses on the effect of the DTMs (e.g. ASTER DTM, ALSM, digitized from paper map and digital photogrammetric measurement data). The main aim of the study is to generate landslide susceptibility index map using the defined relationship using ANN. Landslide data in the Chuetsu region were used in this research. The 2004 earthquake triggered many landslides in the region. The initial results of the study showed that ANN is more accurate in defining the relationship between geological and geomorphological factors and landslide occurrence. It also determined the best combination of geological and geomorphological factors that is directly related to landslide occurrence.
DE: 1826 Geomorphology: hillslope (1625)
DE: 1849 Numerical approximations and analysis
DE: 3252 Spatial analysis (0500)
DE: 9320 Asia
SC: Hydrology [H]
MN: 2007 Fall Meeting