Application and cross-validation of spatial logistic multiple regression for landslide susceptibility analysis

摘要

The aim of this study is to apply and cross-validate a spatial logistic multiple-regression model at Boeun, Korea using a Geographic Information System (GIS). For this landslide locations in the Boeun area were identified by interpretation of aerial photographs and field surveys. Maps of the topography, soil type, forest cover, geology and land cover were constructed from a spatial database. The 13 factors that influence landslide occurrence were calculated and extracted from the spatial database. Using the 13 factors, landslide susceptibility was analyzed by logistic multiple-regression methods. For validation and cross-validation, the result of the landslide susceptibility analysis obtained from Boeun area was applied to Yongin area in Korea. The validation and cross-validation results showed 75.0% and 85.3% prediction accuracy between the susceptibility map and the existing landslide locations. The GIS was used to analyze the vast amount of data efficiently and statistical programs were used to maintain specificity and accuracy.