Abstract
Susceptibility mapping provides information about vulnerable locations and thus helps to potentially decrease infrastructure damage due to mass wasting. During the past decades, expansion of settlements into areas prone to landslides in Iraq has highlighted the importance of accurate landslide susceptibility studies. The main goal of this research is to implement selected morphometric parameters to improve prediction of landslide susceptibility in the Zagros Mountain region. We used the Mawat area, in the Kurdistan Region (NE Iraq) to test the added value of morphometric indicators. Sixteen morphometric factors, mainly derived from a Digital Elevation Model (DEM), extracted using the stereo-ability of the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) satellite, as well as geological and environmental predictive factors, were appraised. We evaluated and compared Frequency Ratio (FR), Weight of Evidence (WOE), Logistic Regression (LR) and Probit Regression (PR) approaches in combination with morphometric indices to determine the Landslide Susceptibility (LS). The areas under the curve (AUC) of the Prediction Rate Curve (PRC), Relative landslide density Index (R index), and True Positive Percentage (TPP) for the four models show that all models perform similarly, and the focus should be on careful selection of the predictive factors, which is far more important than the methods used. Results indicate that lithology and slope aspects are the more dominant factors that lead to detect possible occurrence of landslides. Furthermore, this work demonstrates that the hypsometric integral performs better than the commonly used slope curvature as a predictor and thus increases the prediction accuracy of the susceptibility map. We argue that the use of adequate morphometric parameters can increase the efficiency of the LS mapping in other regions of the world.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have