Figures (8)  Tables (3)

    • Figure 1. 

      Location of the 2013 Ms6.6 Min-Zhang earthquake.

    • Figure 2. 

      Model framework for seismic landslide prediction.

    • Figure 3. 

      Seismic landslide inventory map.

    • Figure 4. 

      Influencing factors used in landslide susceptibility mapping. (to be continued)

    • Figure 4. 

      (continued).

    • Figure 4. 

      (continued).

    • Figure 5. 

      ROC curve for model testing.

    • Figure 6. 

      Results of the predicted landslide susceptibility.

    • ReferenceYearInfluencing factorsMethodology
      Gigovic et al.[13]2019Elevation, slope, aspect, distance to road network, distance to river, distance to fault, lithology, NDVI, terrain wetness index (TWI), stream power index (SPI), sediment transport index (STI), annual rainfall, distance to urban area, and land use rateMulti-criteria decision analysis best–worst methodology
      Xu et al.[14]2012Slope, aspect, slope curvature, elevation, surface catchment, drainage distance, road distance, lithology, fault distance, and NDVIWeight of evidence modeling
      Xu et al.[19]2012Elevation, slope angle, slope direction, fault distance, river distance, and lithologySupport vector machine
      Yao et al.[20]2008Lithology, slope angle, slope aspect, elevation and vegetationSupport vector machine
      Ma & Xu[21]2019Permanent displacement, critical acceleration, terrain, peak ground acceleration, river distance, and road distanceNewmark model, support vector machine, logic regression

      Table 1. 

      Overview of selection of influencing factors and methodology.

    • Influencing factorTypeResolutionData sourceYear updated
      PGARaster30 m × 30 mInstitute of Engineering Mechanics, China Earthquake Administration2013
      Epicenter distanceRaster30 m × 30 mInternet2013
      ElevationRaster30 m × 30 mGeospatial Data Cloud site, Computer Network Information Center, Chinese Academy of Sciences2009
      SlopeRaster30 m × 30 mExtracted from DEM2009
      AspectRaster30 m × 30 mExtracted from DEM2009
      Plan curvatureRaster30 m × 30 mExtracted from DEM2009
      Profile curvatureRaster30 m × 30 mExtracted from DEM2009
      Fault distanceRaster1:100000China Earthquake Network Center2011
      River distanceRaster1:100000China Earthquake Network Center2011
      NDVIRaster90 m × 90 mGeospatial Data Cloud site, Computer Network Information Center, Chinese Academy of Sciences2012

      Table 2. 

      Data details of influencing factors.

    • ModelR2AUC
      KNN0.9880.986
      NB0.8110.988
      RF0.9950.999
      LR0.9890.997
      SVM0.9900.998

      Table 3. 

      Prediction performance for KNN, NB, RF, LR, and SVM.