Ayub Mohammadi

Among the natural hazards, landslide is considered as one of the complex hazards and at the same time harmful, as a result of their appearance, slope masses such as mud and pieces of rocks are moved from the sloped surfaces and cause problems on the slopes. puts Every year, this phenomenon causes a lot of damage to residential areas, roads, etc., both financially and personally. The purpose of this research is to identify and zonate the risk of landslides in Baneh-Marivan mountain road in Kurdistan province using satellite images and advanced data mining algorithms. Factors affecting the occurrence of landslides that were considered in this study include slope, slope direction, slope curvature, height, distance from waterway, distance from road, distance from fault, density of fault, density of waterway, lithology, rainfall, land use, Topographic humidity index, current power index and vegetation density index. After collecting the required information and layers, 126 slip points were identified using satellite images. Landslide situations were randomly divided into two groups of 70% (88) and 30% (38) for modeling and validation, respectively. Landslide zoning map of Baneh-Marivan road was prepared with the help of EBF and Random Forest (RF) models. To evaluate the performance of the models, the area under the AUC graph obtained from the ROC curve was used. According to the evaluation criteria used in this study (ROC) and according to the validation data, the random forest (RF) model with AUC (0.907) had the best performance in landslide detection in the target area and after The EBF model had the lowest performance in identifying landslides (0.881) in the study area. Finally, the results of the study showed that the landslides that occurred on the Baneh-Marivan road are caused by environmental factors. It can also be acknowledged that the landslide risk prediction map provided by advanced data mining algorithms can help managers of natural hazards in different organizations as a suitable scientific solution.

Landslides are natural disasters that cause a lot of financial and life losses in the country, annually. Identifying high risk areas can reduce the damages and be effective on land development policies. The aim of this study was to maping the landslide hazard of Sanandaj-Marivan road in Kurdistan province. In first step, the technique used in this study was differential artificial valve interferometry. 44 landslides were detected in the study area using radar images. In order to check and validate the final map of landslide distribution points, Google Earth software was used. The location of the landslides detected was also investigated using GPS and field navigation and another 29 landslides were identified according to field studies. In the second step, advanced data mining algorithms including evidential belief function (EBF), weighted evidence (WOE) and support vector machine (SVM) were used to landslide risk mapping. Firstly, 73 locations of landslides were categorized into two groups of 70% (51 locations) and 30% (22 locations), randomly, for training and validation processes, respectively. Totally, 14 landslide conditioning factors including slope, aspect, elevation, distance to river, distance to road, river density, distance to fault, rain, normalized difference degetation index, land use, slope curvature, lithology, stream power index (SPI) and topographic wetness index (TWI) were obtained from various data sources to landslide hazard mapping. Subsequently, the resulting three models were validated and compared using some statistical metrics including area under the receiver operating characteristics (AUROC) curve. The AUC results introduced the success rates of 0.538, 0.717 and 0.97 for EBF, WOE and SVM, respectively. Therefore, the SVM model, having the highest AUC, was the most accurate method among the three implemented methods in identifying the regions at risk of future landslides in the study area. In general, it can be said that a rigorous spatial forecasting map can help managers and urban planners in identifying landslide sensitive areas for disaster management.

Landslides are among the most complex and at the same time harmful phenomena that due to their occurrence and emergence, masses of slopes are displaced from sloping surfaces and leave effects on the surface of slopes. Occurrence of such phenomena near tolls such as residential areas, roads, etc. causes a lot of financial and human losses. The purpose of this study is to identify and zoning landslide hazard on the mountainous road Tahm-Chorzeq in Zanjan province using satellite images and advanced data mining algorithms. Factors influencing landslide occurrence considered in this study include precipitation, slope curvature, slope direction, slope, height, distance from road, distance from river, distance from fault, waterway density, lithology, soil texture, land use, TWI, SPI, and NDVI. After collecting the required information and layers, 38 landslide points were identified using satellite images, then by land survey of these 38 points, 25 The point was confirmed and the rest of the points were removed and in the same survey new points were collected and the total points were identified to 150 points. Landslide positions were randomly divided into two groups of 70% (105) and 30% (45), respectively. Divided for modeling and validation. Using the models of definitive evidence functions (EBF), statistical index (SI) and weighted evidence (WOE), the landslide map of Tham-Chorzeq road was prepared. In order to evaluate the performance of the models, the area below the AUC diagram obtained from the ROC curve was used. According to the evaluation criteria used in this study (ROC) and according to the validation data, the conclusive evidence model EBF (0.850) had the best performance in landslide potential in the area and then to The SI statistical index model (0.740) and the WOE weighted evidence model (0.704) had the lowest performance in landslide potential detection in the study area, respectively. Finally, the results of the study showed that landslides on the Tham-Chorzeq road due to environmental factors include land use change and distance from the road. It can also be acknowledged that landslide risk prediction maps provided by advanced data mining algorithms can be a great scientific solution to help natural disaster managers in various organizations.

زمین لغزش ها از جمله بلایای طبیعی هستند که سالانه خسارت های مالی و جانی زیادی را در کشور ایجاد می کنند. شناخت مناطق پرخطر می تواند در کاهش خسارت ها و تصمیم گیری در سیاست های توسعه اراضی مؤثر باشد. هدف این مطالعه، پهنه بندی خطر زمین لغزش محدوده جاده ارتباطی سنندج-کامیاران در استان کردستان بود. تکنیک مورد استفاده در این پژوهش، تداخل سنجی دریچه مصنوعی تفاضلی بود. در این مطالعه با استفاده از دو تصویر SAR و یک تصویر DEM، تداخل نگارها ایجاد شدند. تصاویر رادار مورد استفاده به صورت SLC سفارش داده شدند. با ارسال پروپوزال به سازمان فضایی اروپا و موافقت این سازمان، داده های راداری فراهم شد. 25 زمین لغزش در منطقه مورد مطالعه با استفاده از تصاویر راداری شناسایی شدند. به منظور بررسی و صحت سنجی نقشه نهایی نقاط پراکنش زمین لغزش از نرم افزار گوگل ارث استفاده شد. همچنین موقعیت زمین لغزش های شناسایی شده با استفاده از GPS و پیمایش میدانی مورد بررسی قرار گرفتند. 25 زمین لغزش در منطقه مورد مطالعه با استفاده از تصاویر راداری شناسایی شدند. در نهایت، روایی و صحت دقیق مکانی نقاط زمین لغزش اتفاق افتاده در طول جاده مورد مطالعه با استفاده از تصاویر گوگل ارث تایید شد. تهیه نقشه زمین لغزش های اتفاق افتاده با استفاده از روش پیمایش میدان به ویژه در مناطق مرتفع، دشوار می باشد. زمین لغزش هایی که در مناطق مرتفع اتفاق می افتند به دلیل شناسایی نشدن، از بین می روند؛ بنابراین توصیه می شود، شناسایی زمین لغزش های اتفاق افتاده با استفاده از تصاویر ماهواره ای با وضوح بالا انجام شود.

Whether they occur due to natural triggers or human activities, landslides lead to loss of life and damages to properties which impact infrastructures, road networks and buildings. Landslide Susceptibility Map (LSM) provides the policy and decision makers with some valuable information. This study aims to detect landslide locations by using Sentinel-1 data, the only freely available online Radar imagery, and to map areas prone to landslide using a novel algorithm of AB-ADTree in Cameron Highlands, Pahang, Malaysia. A total of 152 landslide locations were detected by using integration of Interferometry Synthetic Aperture RADAR (InSAR) technique, Google Earth (GE) images and extensive field survey. However, 80% of the data were employed for training the machine learning algorithms and the remaining 20% for validation purposes. Seventeen triggering and conditioning factors, namely slope, aspect, elevation, distance to road, distance to river, proximity to fault, road density, river density, Normalized Difference Vegetation Index (NDVI), rainfall, land cover, lithology, soil types, curvature, profile curvature, Stream Power Index (SPI) and Topographic Wetness Index (TWI), were extracted from satellite imageries, digital elevation model (DEM), geological and soil maps. These factors were utilized to generate landslide susceptibility maps using Logistic Regression (LR) model, Logistic Model Tree (LMT), Random Forest (RF), Alternating Decision Tree (ADTree), Adaptive Boosting (AdaBoost) and a novel hybrid model from ADTree and AdaBoost models, namely AB-ADTree model. The validation was based on area under the ROC curve (AUC) and statistical measurements of Positive Predictive Value (PPV), Negative Predictive Value (NPV), sensitivity, specificity, accuracy and Root Mean Square Error (RMSE). The results showed that AUC was 90%, 92%, 88%, 59%, 96% and 94% for LR, LMT, RF, ADTree, AdaBoost and AB-ADTree algorithms, respectively. Non-parametric evaluations of the Friedman and Wilcoxon were also applied to assess the models’ performance: the findings revealed that ADTree is inferior to the other models used in this study. Using a handheld Global Positioning System (GPS), field study and validation were performed for almost 20% (30 locations) of the detected landslide locations and the results revealed that the landslide locations were correctly detected. In conclusion, this study can be applicable for hazard mitigation purposes and regional planning.