LI Kai, XU Chong, TAN Ming, CHEN Jianbo, LI Shuai, LUO Ju, WU Guodong. 2019: GIS-BASED AND LOGISTIC REGRESSION MODEL FOR HAZARD ASSESS-MENT OF LANDSLIDES TRIGGERED BY HEJING EARTHQUAKE IN 2012. JOURNAL OF ENGINEERING GEOLOGY, 27(s1): 262-268. DOI: 10.13544/j.cnki.jeg.2019102
    Citation: LI Kai, XU Chong, TAN Ming, CHEN Jianbo, LI Shuai, LUO Ju, WU Guodong. 2019: GIS-BASED AND LOGISTIC REGRESSION MODEL FOR HAZARD ASSESS-MENT OF LANDSLIDES TRIGGERED BY HEJING EARTHQUAKE IN 2012. JOURNAL OF ENGINEERING GEOLOGY, 27(s1): 262-268. DOI: 10.13544/j.cnki.jeg.2019102

    GIS-BASED AND LOGISTIC REGRESSION MODEL FOR HAZARD ASSESS-MENT OF LANDSLIDES TRIGGERED BY HEJING EARTHQUAKE IN 2012

    • On June 30, 2012, a large number of co-seismic landslides were triggered by the MS6.6 earthquake at the junction of Hejing County and Xinyuan County in Xinjiang. 306 landslides were interpreted and delineated from the IKONOS remote sensing image whose resolution is 1 m after the earthquake. Most of the landslides were small, Elevation, earthquake intensity, slope, distance to fault, curvature, aspect, slope position, distance to roads, distance to drainage sand lithology were selected as the influencing factors of the earthquake landslides. Rasterized the layers of influence factors on the platform of Geographic Information System(GIS),the aim of this study was to apply logistic regression model for the Hejing earthquake triggered landslide hazard mapping, and then divided research area into extremely high hazard zone, high hazard zone, middle hazard zone, low hazard zone, extremely low hazard zone, and the map of landslide hazard assessment was obtained. The ROC curve was used to validate the prediction results. The result shows that the AUC value of the training sample set is 0.80,which indicates that the accuracy of the LR model is ideal. The final conclusion in the paper provides a reference for disaster prevention and mitigation of landslides in the earthquake zone.
    • loading

    Catalog

      /

      DownLoad:  Full-Size Img  PowerPoint
      Return
      Return