Landslide Susceptibility Modeling Using Geospatial Analysis in Fairbanks North Star Borough, Alaska

ABSTRACT A landslide susceptibility map delineates the potential zones for landslide occurrence, which is critical in landslide management. The aim of this study is to produce a landslide susceptibility model of part of the Fairbanks North Star Borough, Alaska, using the analytical hierarchy process (AHP) and the frequency ratio (FR) methods in a geographical information systems (GIS) environment. To produce the susceptibility map, seven landslide-influencing factors—geology, slope angle, slope aspect, plan curvature, profile curvature, distance to the nearest road, and average annual precipitation—were selected. Spatial databases and thematic layers of all the aforementioned factors were created in a GIS environment. The AHP was applied to derive factor weights that indicated their relative importance in causing landslides. The FR method was used to derive class ranks that indicated the relative importance of individual classes for each factor. All the weighted raster maps of the factors were integrated using a weighted linear combination method to generate a landslide susceptibility map, which classified the study area into five different landslide susceptible zones, i.e., very low, low, moderate, high, and very high. The susceptibility map was validated using the area under curve (AUC) and landslide density index (LDI) methods. The AUC value of the prediction rate curve was calculated, and a 76.8 percent validation accuracy was achieved. The LDI result also indicated a successful landslide susceptibility model. The results of this study can be used to identify slopes prone to sliding on a relative basis.