Search

Landslides lead to a great threat to human life and property safety. The delineation of landslide-prone areas achieved by landslide susceptibility assessment plays an important role in landslide management strategy. Selecting an appropriate mapping unit is vital for landslide susceptibility assessment. This paper compares the slope unit and grid cell as mapping unit for landslide susceptibility assessment. Grid cells can be easily obtained and their matrix format is convenient for calculation. A slope unit is considered as the watershed defined by ridge lines and valley lines based on hydrological theory and slope units are more associated with the actual geological environment. Using 70% landslide events as the training data and the remaining landslide events for verification, landslide susceptibility maps based on slope units and grid cells were obtained respectively using a modified information value model. ROC curve was utilized to evaluate the landslide susceptibility maps by calculating the training accuracy and predictive accuracy. The training accuracies of the grid cell-based susceptibility assessment result and slope unit-based susceptibility assessment result were 80.9 and 83.2%, and the prediction accuracies were 80.3 and 82.6%, respectively. Therefore, landslide susceptibility mapping based on slope units performed better than grid cell-based method. [ABSTRACT FROM AUTHOR]

In this paper, we clustered and analyzed landslides and investigated their underlying driving forces at two levels, country and cluster, all over Iran. Considering 12 conditioning factors, the landslides were clustered into nine relatively homogeneous regions using the Contextual Neural Gas (CNG) algorithm. Next, their underlying driving forces were ranked using the Random Forest (RF) algorithm at country and cluster levels. Our results indicate that the mechanisms for landslide occurrence varied for each cluster and that driving forces of the landslides operated differently at a country level compared to the cluster level. Moreover, slope, altitude, average annual rainfall, and distance to the main roads were identified as the most important causes of landslides within all clusters. Thus, for effective management and modelling landslides on a large scale, the variation in the functionality of effective factors should be considered. [ABSTRACT FROM AUTHOR]

Landslide is a natural hazard that, next to earthquakes and floods, causes the largest damage to humans. Nowadays, experts take landslide events into serious consideration and they try to map and identify them accurately in time domain. So far, most of the methods used for this purpose have been of the time-domain type with their own drawbacks. Dealing with some functions in a time-domain method is difficult due to the complexity of calculations, frequency behavior analysis and content extraction. Therefore, a frequency-domain format was used in this study. The short-time Fourier transform and weighted overlay methods were applied in order to detect landslides-prone areas, and the results were evaluated using the prediction-area plot. As the intersection point of the P–A plot in the power spectrum-area model demonstrated 82% of landslide events had occurred only in 18% of the district, while the intersection point of the P–A plot in the weighted overlay model showed the occurrence of 70% of landslide events only in 30% of the study area. [ABSTRACT FROM AUTHOR]

In recent years, increasing efforts have been made to predict the time, location, and magnitude of future landslides. This study explores the potential application of four state-of-the-art data mining models (logistic regression, random forest, support vector machine, and Naïve Bayes tree) for the spatially explicit prediction of landslide susceptibility across a landslide-prone landscape in the Zagros Mountains, Iran. Fifteen conditioning factors and 272 historical landslide events were used to develop a geospatial database for the study area. A two-step factor analysis procedure based on the multicollinearity analysis and the Gain Ratio technique was performed to measure the predictive utility of the factors and to quantify their contribution to landslide occurrences across the study region. Once the models were successfully trained and validated using several performance metrics (i.e., ROC-AUC, sensitivity, specificity, accuracy, RMSE, and Kappa), they were applied to the entire study region to generate distribution maps of landslide susceptibilities. Overall, the random forest model demonstrated the highest training performance (AUC = 0.971; accuracy = 99%; RMSE = 0.120) and ability to predict future landslides (AUC = 0.942; accuracy =87%; RMSE = 0.312), followed by the support vector machine, Naïve Bayes tree, and logistic regression models. The Wilcoxon signed-rank test further proved the superiority of the random forest model for mapping landslide susceptibility in the Zagros region. The insights obtained from this research could be useful for the spatially explicit assessment of landslide-prone landscapes and obtaining a better understanding of the capability of different predictive models. [ABSTRACT FROM AUTHOR]