Earthquake-induced Landslides Hazard Zonation of Sarpol-e Zahab using Entropy Shannon Model

Earthquake of November 11, 2017 (Mw=7.3) in Ezgeleh village of Sarpol-e Zahab city in Iran, triggering numerous slope instabilities of various types of rockfalls, rockslides, avalanches and mud flows. The existence of a very high inherent potential and susceptibility of the region, on the one hand, and the occurrence of a strong earthquake as a driving factor, on the other hand, have been the main reasons for the instabilities in the region have been the main reasons for the instabilities in the region. The number of slope instabilities caused by this earthquake was rare compared to earthquakes with similar magnitude and thus requires more detailed investigations. The main objective of this study is to make a landslide hazard zonation map in this region, using Entropy Shannon’s model, and compare the results with the landslides that triggered by the Ezgeleh earthquake. Preparing a map can identify and distinct low-risk and high-risk areas regarding soil and rock slope instabilities. The earthquake showed that such a map can help significantly to the future construction activities in the region and mitigate the loss of lives and properties. The landslide conditioning factors such as Slope Angle, Lithology (geology), Geological Strength Index (GSI), Slope Aspect, Distance to Faults, Pick Ground Acceleration (PGA), Plan Curvature, Distance to Roads, Distance to Rivers, Land Use, Normalized Difference Vegetation Index (NDVI) and Topographic Wetness Index (TWI), were extracted from the spatial database. By using these factors, weights of each factor were analyzed by index of Entropy model and the map of landslide hazard zonation were prepared, using Geographical Information System (GIS). The results showed that more than 31.37% of the surface of the area has a moderate to very high hazard of landslides. From 335 landslides identified, 235 (≈ 70%) locations were used for the landslide susceptibility maps, while the remaining 100 (≈ 30%) cases were used for model validation. Finally, the ROC (receiver operating characteristic) curve for landslide hazard zonation map was drawn and the areas under the curve (AUC) were calculated. The verification results showed that the index of Entropy model (AUC = 84.3%) has a high accuracy that is assumed as very good.