Landslide susceptibility mapping in the region of eastern Himalayan syntaxis, Tibetan Plateau, China: a comparison between analytical hierarchy process information value and logistic regression-information value methods

• Published in: Bulletin of engineering geology and the environment Vol. 78; no. 6; pp. 4201 - 4215
• Main Authors: Du, Guoliang, Zhang, Yongshuang, Yang, Zhihua, Guo, Changbao, Yao, Xin, Sun, Dongyan
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.09.2019; Springer Nature B.V
• Subjects: Aerial photographs; Aerial photography; Aerial surveys; Analytic hierarchy process; Area; Conditioning; Curvature; Decision making; Distance; Earth; Earth and Environmental Science; Earth Sciences; Foundations; Geoecology/Natural Processes; Geoengineering; Geographic information systems; Geographical information systems; Geological engineering; Geotechnical Engineering & Applied Earth Sciences; Hydraulics; Information processing; Information systems; Landslides; Landslides & mudslides; Lithology; Mapping; Mathematical analysis; Methods; Nature Conservation; Original Paper; Poverty; Regression analysis; Remote sensing; Satellite navigation systems; Sensitivity analysis; Slopes; Surveys; Uplift; Tibetan Plateau; China; GIS; Landslide susceptibility; Eastern Himalayan syntaxis; Analytical hierarchy process information value; Logistic regression-information value
• ISSN: 1435-9529; 1435-9537
• DOI: 10.1007/s10064-018-1393-4

The eastern Himalayan syntaxis in Tibet is one of the regions tectonically most active with the fastest uplift rate on the earth, where landslides are extremely frequent, causing severe damage to lives and transportation and inducing poverty. Thus, mapping landslide susceptibility of this area is of great importance. The purpose of this study is to compare landslide susceptibility maps for this region produced by the analytic hierarchy process information value (AHPIV) and logistic regression-information value (LRIV) methods using geographic information system (GIS) software. To do this, an inventory map with 799 landslides was prepared based on historical documents, interpretation of aerial photographs, and extensive field surveys. A total of eight conditioning factors were analyzed as input variables: lithology, slope gradient, slope aspect, elevation, curvature, distance to faults, distance to drainages and distance to roads. Then, the AHPIV and LRIV methods were applied to mapping landslide susceptibility. The performances of the methods were validated and compared using receiver operating characteristics (ROC) curves. The area under the curve (AUC) values obtained using the AHPIV and LRIV methods were 0.884, and 0.906, respectively. Results showed that the LRIV method performs better than the AHPIV method. Finally, sensitivity analyses were performed to examine the effects of removing any of the conditioning factors on the landslide susceptibility mapping. Results indicate that all of the conditioning factors have a positive effect on the landslide susceptibility mapping. Therefore, the LRIV method with eight conditioning factors was employed to determine potential landslide zones in the study area for landslide management and decision making.