Search

Globally, and in China, landslides constitute one of the most important and frequently encountered natural hazard events. In the present study, landslide susceptibility evaluation was undertaken using novel ensembles of bivariate statistical-methods-based (evidential belief function (EBF), statistical index (SI), and weights of evidence (WoE)) kernel logistic regression machine learning classifiers. A landslide inventory comprising 222 landslides and 15 conditioning factors (slope angle, slope aspect, altitude, plan curvature, profile curvature, stream power index, sediment transport index, topographic wetness index, distance to rivers, distance to roads, distance to faults, NDVI, land use, lithology, and rainfall) was prepared as the spatial database. Correlation analysis and selection of conditioning factors were conducted using multicollinearity analysis and classifier attribute evaluation methods, respectively. The receiver operating characteristic curve method was used to validate the models. The areas under the success rate (AUC_T) and prediction rate (AUC_P) curves and landslide density analysis were also used to assess the prediction capability of the landslide susceptibility maps. Results showed that the EBF-KLR hybrid model had the highest predictive capability in landslide susceptibility assessment (AUC values of 0.814 and 0.753 for the training and validation datasets, respectively; AUC_T value of 0.8511 and AUC_P value of 0.7615), followed in descending order by the SI-KLR and WoE-KLR hybrid models. These findings indicate that hybrid models could improve the predictive capability of bivariate models, and that the EBF-KLR is a promising hybrid model for the spatial prediction of landslides in susceptible areas. [ABSTRACT FROM AUTHOR]

The aim of this research is to investigate multi-criteria decision making [spatial multi-criteria evaluation (SMCE)], bivariate statistical methods [frequency ratio (FR), index of entropy (IOE), weighted linear combination (WLC)] and machine learning [support vector machine (SVM)] models for estimating landslide susceptibility at the Wuning area, China. A total of 445 landslides were randomly classified into 70% (311 landslides) and 30% (134 landslides) to train and validate landslide models, respectively. Fourteen landslide conditioning factors including slope angle, slope aspect, altitude, topographic wetness index, stream power index, sediment transport index, soil, lithology, NDVI, land use, rainfall, distance to road, distance to river and distance to fault were then studied for landslide susceptibility assessment. Performances of five studied models were evaluated using area under the ROC curve (AUROC) for training (success rate curve) and validation (prediction rate curve) datasets, statistical-based measures and tests. Results indicated that the area under the success rate curve for the FR, IOE, WLC, SVM and SMCE models was 88.32%, 82.58%, 78.91%, 85.47% and 89.96%, respectively, demonstrating that SMCE could provide the higher accuracy. The prediction capability findings revealed that the SMCE model (AUC = 86.81%) was also the highest approach among the five studied models, followed by the FR (AUC = 84.53%), the SVM (AUC = 81.24%), the IOE (AUC = 79.67%) and WLC (73.92%) methods. The landslide susceptibility maps derived from the above five models are reasonably accurate and could be used to perform elementary land use planning for hazard extenuation. [ABSTRACT FROM AUTHOR]

The main goal of this study is to produce landslide susceptibility maps for the Qianyang County of Baoji city, China, using both certainty factor (CF) and index of entropy (IOE) models. At first, a landslide inventory map was prepared using earlier reports and aerial photographs as well as by carrying out field surveys. A total of 81 landslide locations were detected. Out of these, 56 (70%) landslides were randomly selected as training data for building landslide susceptibility models and the remaining 25 (30%) were used for the validation purposes. Then, a total number of 15 landslide causative factors, such as slope angle, slope aspect, general curvature, plan curvature, profile curvature, altitude, distance to faults, distance to rivers, distance to roads, the sediment transport index (STI), the stream power index (SPI), the topographic wetness index (TWI), geomorphology, lithology, and rainfall, were used in the analysis. The susceptibility maps produced using CF and IOE models had five different susceptibility classes such as very low, low, moderate, high, and very high. Finally, the output maps were validated using the validation data (i.e., 30% landslide location data that was not used during the model construction), using the area under the curve (AUC) method. The 'success rate' curve showed that the area under the curve for CF and IOE models were 0.8433 (84.33%) and 0.8227 (82.27%) accuracy, respectively. Similarly, the validation result showed that the susceptibility map using CF model has the higher prediction accuracy of 82.32%, while for IOE model it was 80.88%. The results of this study showed that the two landslide susceptibility maps obtained were successful and can be used for preliminary land use planning and hazard mitigation purpose. [ABSTRACT FROM AUTHOR]

This study assesses the landslide susceptibility of the Youfang area, China. For this purpose, four advanced artificial intelligence models, namely, Naïve Bayes (NB), multilayer perceptron (MLP), kernel logistic regression (KLR), and J48-bagging methods, were applied and compared. The relationship between landslides happening and landslide conditioning factors which include: slope, aspect, altitude, plan curvature, profile curvature, stream power index (SPI), topographic wetness index (TWI), sediment transport index (STI), landuse, lithology, distance to faults, distance to roads, distance to rivers, and rainfall were analyzed by the frequency ratios method. These results indicated that MLP model exhibits the most stable and reasonable result, and the resultant landslide susceptibility maps are a useful tool for local government managers and policy planners for this study area and other areas. [ABSTRACT FROM AUTHOR]