![Sample our Environment & Agriculture journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5934/TOC-Subject-Sample-2021-Environment-Agriculture.jpg"})
Abstract
The purpose of this study was to develop landslide susceptibility analysis techniques using artificial neural networks and then to apply these to the selected study areas of Gangneung in Korea. We aimed to verify the effect of data selection at training sites. Landslide locations were identified by the change detection technique of Korea Multipurpose Satellite (KOMPSAT‐1) electro‐optical camera (EOC) images, checked in the field, and a spatial database of topography, soil, forest, and land use was constructed. Sixteen landslide‐related factors were extracted. These factors were used with an artificial neural network to analyse landslide susceptibility. Each factor weight was determined by the back‐propagation training method. Five different data sets were applied to analyse and verify the effect of training. Landslide susceptibility indices were then calculated using the trained back‐propagation weights, and susceptibility maps were constructed from Geographic Information System (GIS) data for five cases of difference training. The landslide susceptibility map was then compared with known landslide locations and verified. The result showed a prediction accuracy varying from 82.72% to 86.10%. The landslide susceptibility map can be used for land use planning to reduce hazards associated with landslides.
Acknowledgements
This research was supported by the Basic Research Project of the Korea Institute of Geoscience and Mineral Resources (KIGAM) funded by the Ministry of Science and Technology of Korea.
