Abstract
The Comparative Research of Landslide Susceptibility Mapping
Track: Teaching with GIS in Higher Education
Author
The landslides in natural slopes generally occur by complicated problems such as soil properties, topography and geology. FR, AHP, LR, ANN is widely used to solve complicated problems in many research fields. The purpose of this study is the development, application and assessment of FR, AHP, LR, ANN methods for assessing landslide susceptibility in the Republic of Korea. Landslide-related factors such as topographical, hydrological, soil, forest, landcover factor were used in the LSI analysis. Each factors was constructed using 3D Analyst, Spatial Analyst(surface, distance, zonal) tools and Analysis tools(extract, overlay, clip, statics) of ArcGIS9.3. LSI maps were produced from FR,AHP,LR,ANN by Analysis tools, spatial analyst tool using constructed input data. And then were compared by means of their validations. Respective AUC values of FR, AHP, LR and ANN were 0.794,0.789,0.794,and 0.806. All model showed a similar accuracy and the map obtained from ANN is more accurate than the other models
Che-Young Oh
Pukyoung National University
599-1, Daeyeon 3-Dong, Nam-Gu
Busan, Busan 608-737
Korea, South
Phone: 82-51-629-6655
Fax: 82-52-629-6653
E-mail: leeieel@nate.com
Chul-Uong Choi
Pukyoung National University
599-1, Daeyeon 3-Dong, Nam-Gu
Busan, Busan 608-737
Korea, South
Phone: 82-51-629-6655
E-mail: cuchoi@pknu.ac.kr
Cheong-Gil Jin
Pukyoung National University
599-1, Daeyeon 3-Dong, Nam-Gu
Busan, Busan 608-737
Korea, South
Phone: 82-51-629-6655
Fax: 82-51-629-6653
E-mail: rekime@nate.com