Geomorphology
Fariba Esfandyari Darabad; Ghobad Rostami; Raoof Mostafazadeh; Mousa Abedini
Abstract
In the current study, the risk of landslides in the Zamkan Watershed, located in Kermanshah Province, was evaluated. Two machine learning models, Support Vector Machine (SVM), and Logistic Regression, were used to prepare a landslide susceptibility map. Toward this, 13 informational layers including ...
Read More
In the current study, the risk of landslides in the Zamkan Watershed, located in Kermanshah Province, was evaluated. Two machine learning models, Support Vector Machine (SVM), and Logistic Regression, were used to prepare a landslide susceptibility map. Toward this, 13 informational layers including elevation, slope, aspect, Melton ruggedness number, terrain convexity, stream length, valley depth, topographic wetness index, precipitation, geological formations, distance from rivers, distance from roads, and vegetation cover were utilized as independent variables. Approximately 70% of the watershed's landslide pixels were used for model training, and 30% for model validation. Model validation was performed using ROC curves. The results indicated the higher performance and accuracy of the radial basis function (RBF) kernel of the SVM model for generating landslide hazard maps in the study area. The area under the curve (AUC) for the RBF kernel was approximately 0.951 for model training and 0.944 for model testing. The results suggest that slope with a coefficient of 0.28, precipitation with a coefficient of 0.27, lithology with a coefficient of 0.26, and elevation with a coefficient of 0.22 are the main controlling factors for landslides occurrence in the Zamkan Watershed. Both the SVM model and logistic regression confirmed the deterministic effects of selected factors on landslides. About 35% of the study area as classified as highly susceptible to landslides, primarily in the eastern half of the watershed. Factors such as high elevation, steep slopes, heavy precipitation, and the Kazhdomi Formation's composition were identified as key contributors to this susceptibility.
Mojgan Entezari; Tahere Jalilian
Volume 6, Issue 18 , June 2019, , Pages 19-38
Abstract
IntroductionLandslide as a natural hazard is very dangerous especially in mountainous areas. It results in loss of human life and property around the world. In spite of the progress in identifying, measuring, predicting, and landslide warning systems, the damage caused by landslides is still increasing ...
Read More
IntroductionLandslide as a natural hazard is very dangerous especially in mountainous areas. It results in loss of human life and property around the world. In spite of the progress in identifying, measuring, predicting, and landslide warning systems, the damage caused by landslides is still increasing worldwide. Therefore, given the importance of the problem, the most important managerial goals include favorable sustainable development in watershed and urban management, and the prediction and controlling of landslide with the aim of reducing its dangers. Indeed, many landslide damages are caused due to not observing correct principles of residential development, dam construction, and construction of roads and facilities. Consequently, the identification of the areas prone to landslide has a great importance for executive organizations. Indeed, the mentioned organizations knowing the location of these areas, they should certainly prevent structure construction in these areas as much as possible. In addition, if it is necessary, they should consider required technical tips and arrangements with more precision. According to the cost of performance, prioritizing the sub basins is very important. Decision making methods is an effective tool to deal with issues that may be created and in this context it has a lot of use. In recent years, attention to the ranking methods in environmental studies have been increased, especially in natural hazards risk management. In this paper, considering the importance and efficiency of the non-ranked ELECTRE-1 method and its non-compensatory nature, we tried to apply this method in the prioritization of landslide risk assessment in six sub-watersheds at Kermanshah province based on the factors and indicators affecting a landslide. The main objectives of the current research were: (1) identifying the main factors affecting the landslides occurrence in the study area, (2) prioritization of the watersheds based on the risk of landslide occurrences, and (3) introducing critical watersheds regarding landslide occurrence.MethodologyThis method, like other decision-making models, is applicable to choosing the best option among others. And like the TOPSIS model, it prioritizes or ranks options by various criteria. In the ELECTRE-1 method, the weight of the criteria should also be calculated for each option.Landslide risk assessment options for the study period are Mahidshat basin, Deira, Kanekabod, Tajrakbadre, Kangir basin, and Chika basin.In general, there are various indicators for assessing the factors affecting the occurrence of landslides in the basins. According to the survey of location of the study area, of various factors affecting the occurrence of landslide, lithological factors, elevation, slope, slope direction, fault density, drainage density, congestion, land use, temperature, precipitation, and slip density were selected as effective factors.-ELECTREmodelFor the first time, it was developed by Roy (1968) in a situation where real criteria and limited privileged relationships were given. Due to the complexity and high volume of computations, the algorithm of the model was programmed in EXCEL software and the values of each step were obtained. Discussion and Conclusion In this research, a multi-criteria decision-making technique was used to map areas susceptible to landslide. To do so, the factors affecting the slope sensitivity to landslide were collected. Then, to apply ELECTERE I technique to rank the sensitivity of the selected sub watersheds to the landslide, the following steps were consecutively taken. 1) The Performance matrix was created to determine the weights of the criteria. 2) The Normalization and Non-normalization matrices were formed. 3) The Harmonious and Inharmonious matrices along with the Coordinated and Uncoordinated effective matrices were obtained. 4) The final Dominance matrix was calculated. The results suggested that among the selected sub watersheds, Mahidasht Rezevand basin ranked the first having the highest vulnerability to landslide occurrence. BadraTjrk and Chika basins respectively ranked the second and the third. Deira and Kanekabod basins shared the forth rank. Finally, Kangir basin was the least likely basin to suffer from landslide incident. The susceptibility maps of the studied basins together with field surveys confirmed the proper application of ELECTRE method for ranking the sub watersheds based on landside risk. Fig 2 indicates that over 36 percent of the landslides have occurred in the high risk area. The proposed method and findings of this study are invaluable for practitioners and future academic studies.
