Search

The Wenchuan earthquake in 2008 induced thousands of geological hazards. Among them, earthquake-induced landslides are extremely disastrous, causing considerable social and economic losses and damage to the ecology and environment. Therefore, it is of great significance to accurately predict the spatial distribution of earthquake-induced landslides. The research in this paper focuses on ten extremely earthquake-stricken areas of the Wenchuan earthquake; landslide hazard data are collected from the study areas, and the terrain information entropy, distance to rivers, distance to faults, distance to roads, lithology, normalized difference vegetation index (NDVI), peak ground acceleration (PGA) and other landslide conditioning factors are extracted. Combined with a deep neural network (DNN), a landslide hazard spatial prediction model is constructed. Through Dataset 1 and Dataset 2, which are generated randomly, the training dataset (70%) and the validation dataset (30%) are divided to verify the robustness and accuracy of the model, and the landslide susceptibility map of the study area is obtained. Moreover, considering the area under curve (AUC), the impacts of different conditioning factors on the landslide hazard prediction model are analyzed. The research results show that the DNN-based landslide hazard spatial prediction model (AUCMean = 93.66%, RecallMean = 85.70%) has a better prediction performance in the training step. From the validation step, the influences of various factors on the spatial prediction model of landslide hazards established in this paper (from high to low) are as follows: distance to faults, lithology, distance to rivers, PGA, terrain information entropy, NDVI, and distance to roads. The landslide hazard point data and landslide susceptibility map are highly consistent, indicating that the application of a deep learning algorithm in hazard susceptibility assessment is effective and can provide a scientific basis for landslide hazard early warning and disaster prevention and mitigation in mountainous areas prone to earthquake disasters. [ABSTRACT FROM AUTHOR]

Landslides are common natural disasters that cause serious damage to ecosystems and human societies. To effectively prevent and mitigate these disasters, an accurate assessment of landslide hazards is necessary. However, most traditional landslide hazard assessment methods rely on static assessment factors while ignoring the dynamic changes in landslides, which may lead to false-positive errors in the assessment results. This paper presents a novel landslide hazard assessment method for the Zagunao River basin, China. In this study, an updated landslide inventory was obtained for the Zagunao River basin using data from interferometric synthetic aperture radar (InSAR) and optical images. Based on this inventory, a landslide susceptibility map was developed using a random forest algorithm. Finally, an evaluation matrix was created by combining the results of deformation rates from both ascending and descending data to establish a hazard level that considers surface deformation. The method presented in this study can reflect recent landslide hazards in the region and produce dynamic assessments of regional landslide hazards. It provides a basis for the government to identify and manage high-risk areas. [ABSTRACT FROM AUTHOR]

Slow-moving landslide-prone areas (SLAs) are unstable objects on the terrestrial surface that can collapse rapidly when provoked by earthquakes, leading to infrastructure damage. It is critical to identify SLAs prior to earthquake events and assess their hazard-causing probabilities when triggered. An assessment approach of earthquake-triggered geohazards is proposed in this paper by combining interferometric synthetic aperture radar (InSAR) derived SLAs and geological and geomorphological factors. Taking the Ms6.8 Luding earthquake, which occurred in the Sichuan Province of southwestern China on September 5, 2022, as an example, 1320 scenes of Sentinel-1 SAR data in western Sichuan were processed using the small baseline subset (SBAS) InSAR technique before the earthquake. After the earthquake, hazard probability assessment was performed in real-time by filtering the SLAs using a spatial analysis technique with geological and geomorphological factors, e.g., real-time peak ground acceleration (PGA), slope, distance to fault (DTF), and distance to the river (DTR) data. The results show that 11 SLAs were classified into high-risk zones. As verified by the Luding co-seismic landslide dataset from visual interpretation of optical remote sensing images, 142 coseismic landslides were triggered by the earthquake in these high-risk regions. In these areas, an ancient landslide, with high pre-earthquake displacement rates (−50 mm/year) on the scarp was reactivated under the Luding earthquake forces. This method can provide a scientific tool for disaster mitigation and rapid response emergency management. [ABSTRACT FROM AUTHOR]

Machine learning models have been widely used in landslide susceptibility prediction. However, landslide multidimensional feature extraction, model generalization ability, and prediction quantification of deep learning are still challenging. This paper proposes a new approach, the landslide density-based bidirectional long short-term memory (LD-BiLSTM) model with multichannel input and an optimized sampling strategy to predict and map landslide susceptibility in active seismic mountainous areas of Sichuan Province, China. First, to ensure the generalization ability of the LD-BiLSTM model, other regions in Sichuan were selected as the model training area independent of the prediction area (Luding County). Multichannel landslide datasets were constructed to extract high-dimensional geospatial features of landslides. Subsequently, the landslide density of each grid cell was utilized as the label for the corresponding input sample. The LD-BiLSTM model was improved by using transfer learning and sampling optimization strategies, which makes our method attenuate the impact of historical landslide inventory deviation on the spatial susceptibility mode compared with the existing DL model, which usually uses landslide objects (LO) as input sample labels. Model performance evaluation results show that the LD-BiLSTM model (precision = 0.903, recall = 0.899, F1-score = 0.901, Area under receiver operating characteristic curve (AUC) = 0.940) outperformed the LO-BiLSTM model (precision = 0.812, recall = 0.815. F1-score = 0.813, AUC = 0.910) in the case areas. Meanwhile, the performance of the LD-BiLSTM model (AUC = 0.9407) significantly outperformed both the information value (IV) (AUC = 0.7207) model and the random forest (RF) (AUC = 0.8116) models in the landslide prediction area (Luding), which confirms that the proposed LD-based method is superior to traditional LO-based methods. Significantly, our approach can effectively extract the spatial distribution of landslides and predict potential landslides in complex high-mountain environments. [ABSTRACT FROM AUTHOR]

Landslides are a serious natural hazard in the world. A map of landslide susceptibility can help to effectively reduce losses. In this paper, a hybrid ensemble technique based on random subspace (RS) and a J48 decision tree named RS–J48T was proposed for landslide susceptibility mapping. This model could enhance the effect of a single classifier significantly and solve the problem of overfitting. Qingchuan County, Sichuan (China) was taken as a study area. A geospatial database which consisted of 640 landslide locations and 12 factors was constructed for this study. The J48 decision tree, artificial neural network (ANN), and other ensemble techniques like AdaBoost and Bagging, were selected for comparison. Receiver operating curves and some statistical indices were used for model validation. The results showed that the RS–J48T model had the better fitting capability (AUC = 0.875), and the best prediction capability (AUC = 0.769) compared to other models. Overall, the novel hybrid model could be a promising way for generating landslide susceptibility maps for other prone areas. [ABSTRACT FROM AUTHOR]

The assessment of earthquake-induced landslide hazards is an important prerequisite for disaster prevention and reduction in tectonic active areas. However, few studies have considered the amplification effect of site and topography on ground motion parameters and made the peak ground acceleration (PGA) correction. Based on Newmark's method, taking Litang County, Sichuan Province, China, as the study area, considering the site amplification effect and topographic amplification effect, this paper carried out the assessment of landslide hazards under the action of occasional earthquakes with an exceedance probability of 10% and rare earthquakes with an exceedance probability of 2%. The results show that the hazard of earthquake-induced landslides is higher in the high slopes of loose rock in the middle of Litang County and the steep rock slopes with large topographic relief in the northeast, and low in the southern plateau and central basins. The site and topographic conditions have a significant effect on the nonlinear amplification of PGA, and the corrected PGA is even magnified by 2–3 times in steep mountains. Compared with the occasional earthquakes, the influence of rare earthquakes on the initiation and movement distance of landslides is remarkably improved. This study can provide a valuable reference for potential earthquake-induced landslide hazard assessment and seismic landslide emergency response in Litang County. [ABSTRACT FROM AUTHOR]

Landslide hazards have attracted increasing public attention over the past decades due to a series of catastrophic consequences of landslide occurrence. Thus, the mitigation and prevention of landslide hazards have been the topical issues. Thereinto, numerous research achievements on landslide susceptibility assessment have been springing up in recent years. In this paper, four benchmark models including best-first decision tree (BFTree), functional tree, support vector machine and classification regression tree (CART) and were integrated with bagging strategy. Then, these bagging-based models were applied to map regional landslide susceptibility in Jiange County, Sichuan Province, China. Fifteen conditioning factors were employed in establishing landslide susceptibility models, respectively, slope aspect, slope angle, elevation, plan curvature, profile curvature, TWI, SPI, STI, lithology, soil, land use, NDVI, distance to rivers, distance to roads and distance to lineaments. Then utilize correlation attribute evaluation method to weigh the contribution of each factor. Finally, the comprehensive performance of various bagging-based models and corresponding benchmark models was evaluated and systematically compared applying receiver operating characteristic curve and area under curve (AUC) values. Results demonstrated that bagging-based ensemble models significantly outperformed their corresponding benchmark models with validation dataset. Among them the Bag-CART model has the highest AUC value of 0.874; however, the AUC value of CART model is only 0.766, which reflected satisfying predictive capacity of integrated models in some degree. The achievements obtained in this study have some reference values for landslides prevention and land resource planning in Jiange County. [ABSTRACT FROM AUTHOR]

Reliable assessment of landslide susceptibility in broad areas of terrain remains challenging due to complex topography and poor representation of randomly selected negative samples. Assessment in broad areas is now primarily based on grid units, which do not have a clear physical meaning like slope units, and their accuracy is not ideal. Nevertheless, the large amount of manual editing, due to the incorrectly generated horizontal and vertical lines during slope unit partitioning, limits using slope units for rapid assessment over large areas. Hence, this paper proposes a reliable susceptibility assessment approach to solve this problem based on optimal slope units and negative samples involving prior knowledge. Precisely, an algorithm to automatically extract slope units is designed to eliminate fragmented and erroneous units. Second, a samples labeling index (SLI) is defined based on the certainty factors model to select negative samples reasonably. Sichuan Province, China is selected for experimental analysis, with the results demonstrate that the optimized slope unit and the negative samples selection strategy consider prior knowledge achieve better results in the random forest model, support vector machine model, and artificial neural network model. In particular, the composite performance index AUC of artificial neural network model improved from 0.81 to 0.90. [ABSTRACT FROM AUTHOR]

Critical acceleration is an inherent property of a slope and determines the slope stability under seismic action. The critical acceleration model is a core element of regional seismic landslide hazard assessment. Therefore, the purpose of this paper is to reveal the influence of different critical acceleration models on assessments of potential earthquake–induced landslide hazards. Traditionally, the Newmark critical acceleration model has commonly been used to evaluate the potential earthquake–induced landslide hazard. This method needs to assume the failure depth of the slope, which leads to an underestimation of the predicted displacement of the seismic landslide. Recently, the prediction equations of critical acceleration based on a parametric study of the limit equilibrium method overcomes the limitation of Newmark critical acceleration model and has been applied to assessments of co-seismic landslide hazards. In this study, we use Newmark critical acceleration model and prediction equations of critical acceleration to obtain the distribution maps of potential earthquake–induced landslide hazard in Shimian County, with peak ground acceleration of 10% and 2% exceeding the probability in 50 years. In addition, the nonlinear effect of site and topographic effects on peak ground acceleration were considered. The results show that Newmark critical acceleration model underestimates the area and value of the predicted displacement, while prediction equations of critical acceleration produces seismic landslides in a wider range of mountainous areas. This indicates that the critical acceleration model has a significant influence on assessments of potential earthquake–induced landslide hazards. In addition, the study not only provides valuable reference for assessment of potential earthquake–induced landslide hazard, emergency response of seismic landslides, and land planning in the study area, it also provides a useful demonstration for the selection of a critical acceleration model in seismic landslide hazard assessments for future researches. [ABSTRACT FROM AUTHOR]

A dataset of landslides from Sichuan Province in China, containing 1551 historical individual landslides, is a result of two teams' effort in the past few years to map the susceptibility to landslides. Considering complex internal relations among the triggering factors, logistic regression (LR) and shallow neural networks, such as back-propagation neural network (BPNN), are often limited. In this paper, we make a straightforward development that the deep belief network (DBN) based on deep learning technology is introduced to map the regional susceptibility to landslides. Seven factors with respect to geomorphology, geology and hydrology are considered and verified through the collinearity test. A DBN model containing three pre-trained layers of restricted Boltzmann machines by stochastic gradient descent method is configured to obtain the susceptibility to landslides. Susceptibility results evaluated by DBN model are compared with those by LR and BPNN in the receive operator characteristic (ROC) analysis, showing that DBN has a better prediction precision, with a lower rate of false alarms and fake alarms. The case study also indicates different sensitivities of the triggering factors to the landslide susceptibility, that the factors of altitude, distance to drainage network and average annual rainfall have significant impact in mapping the susceptibility to landslides in the region. This research will contribute to a better-performance model for regional-scale mapping for the susceptibility to landslides, in particular, at the area where triggering factors show complex relations and relative independence. [ABSTRACT FROM AUTHOR]

Landslides are the most common and destructive secondary geological hazards caused by earthquakes. It is difficult to extract landslides automatically based on remote sensing data, which is import for the scenario of disaster emergency rescue. The literature review showed that the current landslides extraction methods mostly depend on expert interpretation which was low automation and thus was unable to provide sufficient information for earthquake rescue in time. To solve the above problem, an end-to-end improved Mask R-CNN model was proposed. The main innovations of this paper were (1) replacing the feature extraction layer with an effective ResNeXt module to extract the landslides. (2) Increasing the bottom-up channel in the feature pyramid network to make full use of low-level positioning and high-level semantic information. (3) Adding edge losses to the loss function to improve the accuracy of the landslide boundary detection accuracy. At the end of this paper, Jiuzhaigou County, Sichuan Province, was used as the study area to evaluate the new model. Results showed that the new method had a precision of 95.8%, a recall of 93.1%, and an overall accuracy (OA) of 94.7%. Compared with the traditional Mask R-CNN model, they have been significantly improved by 13.9%, 13.4%, and 9.9%, respectively. It was proved that the new method was effective in the landslides automatic extraction. [ABSTRACT FROM AUTHOR]

Landslide susceptibility assessment (LSA) is an essential tool for landslide hazard warning. The selection of earthquake-related factors is pivotal for seismic LSA. In this study, Newmark displacement (Dn) is employed as the earthquake-related factor, providing a detailed representation of seismic characteristics. On the algorithmic side, a dual-channel convolutional neural network (CNN) model is built, and the last classification layer is replaced with two machine learning (ML) models to facilitate the extraction of deeper features related to landslide development. This research focuses on Beichuan County in Sichuan Province, China. Fifteen landslide predisposing factors, including hydrological, geomorphic, geological, vegetation cover, anthropogenic, and earthquake-related features, were extensively collected. The results demonstrate some specific issues. Dn outperforms conventional earthquake-related factors such as peak ground acceleration (PGA) and Arias intensity (Ia) in capturing seismic influence on landslide development. Under the same conditions, the OA improved by 5.55% and AUC improved by 0.055 compared to the PGA; the OA improved by 3.2% and AUC improved by 0.0327 compared to the Ia. The improved CNN outperforms ML models. Under the same conditions, the OA improved by 4.69% and AUC improved by 0.0467 compared to RF; the OA improved by 4.47% and AUC improved by 0.0447 compared to SVM. Additionally, historical landslides validate the reasonableness of the landslide susceptibility maps. The proposed method exhibits a high rate of overlap with the historical landslide inventory. The proportion of historical landslides in the very high and high susceptibility zones exceeds 87%. The method not only enhances accuracy but also produces a more fine-grained susceptibility map, providing a reliable basis for early warning of seismic landslides. [ABSTRACT FROM AUTHOR]

Calculating the failure probability of a landslide is important in engineering related to geological process and geomorphological evolution. Strength parameters of a soil (i.e., cohesion and internal friction angle) are regarded as uncertain-but-bounded parameters. In this study, a new method is proposed for computing the landslide failure probability based on a convex set model and artificial neural network (ANN). In the new method, ANN is used to determine the limit state function of landslide stability, and the failure probability is determined using a simple iterative algorithm. The new method was applied to calculate the failure probability of the Gufenping landslide in Nanjiang, Sichuan, China. The results calculated using a Monte Carlo simulation (MCS) method confirmed that the new method accurately and quickly obtains the failure probability of a landslide. Additionally, compared with the two-dimensional calculation method, the one-dimensional analysis method overestimates the failure probability of the landslide. The results of single factor and global sensitivity analysis indicate that the average of internal friction angle is the main factor affecting the stability of the landslide. It is easy to calculate failure probability of landslides using the novel method than using the conventional methods. [ABSTRACT FROM AUTHOR]

The purpose of this work is to carry out seismically induced landslide probabilistic hazard mapping for future seismic scenarios of Aba Prefecture and Chengdu Plain region, Sichuan Province, China. Nine earthquake events that occurred in the regions and neighboring areas are selected, which include a total of 251,260 landslide records. This work used 12 influencing factors including elevation, slope, aspect, relief, topographic wetness index (TWI), topographic position index (TPI), peak ground motion, distance to active faults, vegetation coverage, distance to roads, lithology, and annual rainfall to establish the LR model. Based on the probabilistic seismic hazard analysis (PSHA) method, the distribution of predicted seismic motion under four earthquake scenarios is calculated including frequent, occasional, rare, and very rare earthquake occurrence. Using the PGA distribution of the four scenarios as input peak ground motion parameters, we calculated the occurrence probability of coseismic landslides in the entire Aba Prefecture and Chengdu Plain region under the action of different ground motions. The result shows that the high-hazard areas are mainly concentrated in the Longmenshan fault zone, and the southern area of Kangding is also a potential high-hazard area for landsliding. Meanwhile, as the probability of exceedance decreases, the probability of corresponding earthquake-induced landslides hazard probability and the area of high-hazard regions also significantly increase. Especially, the Pengguan complex rock mass in the southwest of the Longmenshan fault zone is the potential high-hazard area for coseismic landslides. Highlights: Establishing a coseismic landslides evaluation model suitable for Aba Prefecture and Chengdu Plain region. Calculating the occurrence probability of coseismic landslides under different peak ground motions. [ABSTRACT FROM AUTHOR]

The region near Dujiangyan City and Wenchuan County, Sichuan China, including significant giant panda habitats, was severely impacted by the Wenchuan earthquake. Largearea landslides occurred and seriously threatened the lives of people and giant pandas. In this paper, we report the development of an enhanced multi-temporal interferometric synthetic aperture radar (MTInSAR) methodology to monitor potential post-seismic landslides by analyzing coherent scatterers (CS) and distributed scatterers (DS) points extracted from multi-temporal L-band ALOS/PALSAR data in an integrated manner. Through the integration of phase optimization and mitigation of the orbit and topography-related phase errors, surface deformations in the study area were derived: the rates in the line of sight (LOS) direction ranged from -7 to 1.5 cm/a. Dozens of potential landslides, distributed mainly along the Minjiang River, Longmenshan Fault, and in other the high-altitude areas were detected. These findings matched the distribution of previous landslides. InSAR-derived results demonstrated that some previous landslides were still active; many unstable slopes have developed, and there are significant probabilities of future massive failures. The impact of landslides on the giant panda habitat, however ranged from low to moderate, would continue to be a concern for conservationists for some time in the future. [ABSTRACT FROM AUTHOR]

On April 20, 2013, a Mw 6.6 earthquake struck Lushan County of Sichuan Province, China, about 80 km southwest of the epicenter of the 12 May 2008 Wenchuan Mw 7.9 event. It triggered a large number of landslides in a broad area, causing causality and injuries and damages to roads, drains, dwellings, and infrastructures. Field investigations of coseismic landslides provide a basis for better understanding and illustrating the spatial distributions and hazards related to the Lushan earthquake-triggered landslides. Therefore, this paper presents results of field investigations of the Lushan earthquake-triggered landslides, which can be classified into two main categories of disrupted landslides and coherent landslides, and constitutes seven types. The disrupted landslides include five types: rock falls, rock slides, rock avalanches, soil falls, and soil slides; and the coherent landslides have two types: soil slumps and slow earth flows. All of the seven types of landslides can be observed in many field photos, aerial photographs, and/or satellite images. Among them, seven typical large-scale landslides, distributed in a northeast trending line consistent with the Shuangshi-Dachuan fault, which is presumably related with the earthquake, are described in detail. We prepared an original, emergency-based coseismic landslide inventory map based on visual interpretation of available post-earthquake aerial photographs and pre-earthquake satellite images, combined with results from emergency-based field investigations. The original inventory map registered 3884 landslides. Later, 11,761 more landslides were identified by much more careful and time-consuming visual interpretation based on the same aerial photographs and satellite images. Subsequently, all the 15,645 coseismic landslides were correlated with six common landslide impact factors, including elevation, slope angle, slope aspect, topographic position, stratum/lithology, and peak ground accumulation (PGA). These observational and compiled coseismic landslide data allow us to make subsequent compiling of a detailed and completed landslide inventory by visual interpretation of remote sensing images as well as further studies of landslides induced by the Lushan earthquake. [ABSTRACT FROM AUTHOR]

The Lushan earthquake with magnitude M 7.0 ( M 6.6, USGS) in Sichuan Province, China, triggered a large number of landslides, which seriously aggravated the earthquake's destructive consequences. This paper mainly focuses on the methodology of the urgent landslide susceptibility assessment right after the earthquake. The detailed landslide inventory (including 5,688 landslides) is prepared by means of urgent post-earthquake landslide field survey, landslide remote sensing interpretation of multi-source remote sensing images including high-resolution unmanned aerial vehicle images and historical landslide archives. Ten remarkable causative factors for landslide occurrence have been selected to conduct the landslide susceptibility assessment, including earthquake intensity, landslide density and slope gradient. An integration assessment approach is developed to facilitate the effective urgent post-earthquake landslide susceptibility assessment using three methods: factor sensitivity analysis, analytical hierarchy process and factor-weighted overlay. Such integration can effectively reduce the subjectivity and uncertainty resulting from using single method. The validation evaluation using the area under curve suggests the landslide susceptibility assessment results have satisfactory accuracy, and the suggested methodology is effective for the urgent post-earthquake landslide susceptibility assessment. The study results reveal that earthquake intensity and slope gradient are the two most important causative factors for post-earthquake landslide occurrence in the Lushan earthquake-impacted area. The dominant slope gradient and slope aspect with relatively higher landslide frequency are 45°-50° and south-east direction, respectively. The intense earthquake impact increased the dominant slope gradient of landslide spatial distribution, and the thrust campaign of seismogenic fault with strike NE-SW made south-east direction as the dominant slope aspect of the landslide spatial distribution. The locations with very high and high landslide susceptibility are mainly distributed in the regions with higher earthquake intensity and adverse terrain conditions, such as Shuangshi town and Longmen town of Lushan county and Muping town of Baoxing county. The study results are expected to provide a beneficial reference for the landslide prevention and infrastructure reconstruction after the Lushan earthquake. [ABSTRACT FROM AUTHOR]

During the five rainy seasons following the 2008 Wenchuan earthquake, at least eight catastrophic events of multiple debris flows occurred in the Yinchanggou area of Baishui River basin, in southwestern China. The deposits had some serious impacts on the transportation, communication, and reconstruction activities. It indicates the start of a period with a high-frequency debris flow activities. In this paper, a detailed review is made of the latest 12 debris flow events, which occurred on 18th August 2012. Field reconnaissance and measurements, supported by a remote sensing interpretation, were conducted to reveal the locations and morphological characteristics of debris flow catchments. A total volume of 50.26 million m co-seismic landslides, triggered by the Wenchuan earthquake, served as source materials for these debris flows and/or shallow landslides. Bedrocks, which were on the hanging wall of the Yingxiu-Beichuan Fault with a mean slope gradient larger than 30°, were the most susceptible for these seism-triggered failures. The volume of debris flow deposits on individual fans vary by many orders of magnitude from 2,900 to 223,500 m. The huge amount of reserves of source material compared to debris flow volumes currently deposited and taking into account the relatively low morphology index values form an indication of the early stage in the development of debris flows and the immense potential for debris flow occurrence in the future. There are two patterns of debris flow initiation which can be distinguished in this study: (a) the channel initiation pattern of concentrated overland flow erosion in the channel beds and (b) the slope initiation pattern which shows the development of new landslides and a transition into debris flows during a heavy rainfall. A catastrophic debris flow in the Haihuiqiao watershed of the Yinchanggou area was selected to show in detail the mechanism of debris flow initiation as a result of intensive erosion in loose debris materials. The critical rainfall threshold in terms of daily precipitation decreased by 77 % compared to the pre-earthquake threshold. A primary rainfall intensity-duration ( ID) threshold curve was constructed for debris flows for an early-warning system and compared with other global rainfall thresholds. The discussion is focused on the initiation mechanisms of debris flows and the assessment of the critical meteorological threshold. [ABSTRACT FROM AUTHOR]

The Ms 7.0 Jiuzhaigou earthquake that occurred on 8 August 2017 triggered hundreds of landslides in the Jiuzhaigou valley scenic and historic-interest area in Sichuan, China, causing heavy casualties and serious property losses. Quick and accurate mapping of post-disaster landslide distribution is of paramount importance for earthquake emergency rescue and the analysis of post-seismic landslides distribution characteristics. The automatic identification of landslides is mostly based on medium- and low-resolution satellite-borne optical remote-sensing imageries, and the high-accuracy interpretation of earthquake-triggered landslides still relies on time-consuming manual interpretation. This paper describes a methodology based on the use of 1 m high-resolution unmanned aerial vehicle (UAV) imagery acquired after the earthquake, and proposes a support vector machine (SVM) classification method combining the roads and villages mask from pre-seismic remote sensing imagery to accurately and automatically map the landslide inventory. Compared with the results of manual visual interpretation, the automatic recognition accuracy could reach 99.89%, and the Kappa coefficient was higher than 0.9, suggesting that the proposed method and 1 m high-resolution UAV imagery greatly improved the mapping accuracy of the landslide area. We also analyzed the spatial-distribution characteristics of earthquake-triggered landslides with the influenced factors of altitude, slope gradient, slope aspect, and the nearest faults, which provided important support for the further study of post-disaster landslide distribution characteristics, susceptibility prediction, and risk assessment. [ABSTRACT FROM AUTHOR]

Landslide is a natural disaster that seriously affects human life and social development. In this study, the characteristics and effectiveness of convolutional neural network (CNN) and conventional machine learning (ML) methods in a landslide susceptibility assessment (LSA) are compared. Six ML methods used in this study are Adaboost, multilayer perceptron neural network (MLP-NN), random forest (RF), naive Bayes, decision tree (DT), and gradient boosting decision tree (GBDT). First, the basic knowledge and structures of the CNN and ML methods, and the steps of the LSA are introduced. Then, 11 conditioning factors in three categories in the Hongxi River Basin, Pingwu County, Mianyang City, Sichuan Province are chosen to build the train, validation, and test samples. The CNN and ML models are constructed based on these samples. For comparison, indicator methods, statistical methods, and landslide susceptibility maps (LSMs) are used. The result shows that the CNN can obtain the highest accuracy (86.41%) and the highest AUC (0.9249) in the LSA. The statistical methods represented by the mean and variance of TP and TN perform more firmly on the possibility of landslide occurrence. Furthermore, the LSMs show that all models can successfully identify most of the landslide points, but for areas with a low frequency of landslides, some models are insufficient. The CNN model demonstrates better results in the recognition of the landslides' cluster region, this is also related to the convolution operation that takes the surrounding environment information into account. The higher accuracy and more concentrative possibility of CNN in LSA is of great significance for disaster prevention and mitigation, which can help the efficient use of human and material resources. Although CNN performs better than other methods, there are still some limitations, the identification of low-cluster landside areas can be enhanced by improving the CNN model. [ABSTRACT FROM AUTHOR]

Landslides are one of the most common geological hazards worldwide, especially in Sichuan Province (Southwest China). The current study's main purposes are to explore the potential applications of convolutional neural networks (CNN) hybrid ensemble metaheuristic optimization algorithms, namely beluga whale optimization (BWO) and coati optimization algorithm (COA), for landslide susceptibility mapping in Sichuan Province (China). For this aim, fourteen landslide conditioning factors were compiled in a spatial database. The effectiveness of the conditioning factors in the development of the landslide predictive model was quantified using the linear support vector machine model. The receiver operating characteristic (ROC) curve (AUC), the root mean square error, and six statistical indices were used to test and compare the three resultant models. For the training dataset, the AUC values of the CNN-COA, CNN-BWO and CNN models were 0.946, 0.937 and 0.855, respectively. In terms of the validation dataset, the CNN-COA model exhibited a higher AUC value of 0.919, while the AUC values of the CNN-BWO and CNN models were 0.906 and 0.805, respectively. The results indicate that the CNN-COA model, followed by the CNN-BWO model, and the CNN model, offers the best overall performance for landslide susceptibility analysis. [ABSTRACT FROM AUTHOR]

Landslide susceptibility mapping (LSM) is an important decision basis for regional landslide hazard risk management, territorial spatial planning and landslide decision making. The current convolutional neural network (CNN)-based landslide susceptibility mapping models do not adequately take into account the spatial nature of texture features, and vision transformer (ViT)-based LSM models have high requirements for the amount of training data. In this study, we overcome the shortcomings of CNN and ViT by fusing these two deep learning models (bottleneck transformer network (BoTNet) and convolutional vision transformer network (ConViT)), and the fused model was used to predict the probability of landslide occurrence. First, we integrated historical landslide data and landslide evaluation factors and analysed whether there was covariance in the landslide evaluation factors. Then, the testing accuracy and generalisation ability of the CNN, ViT, BoTNet and ConViT models were compared and analysed. Finally, four landslide susceptibility mapping models were used to predict the probability of landslide occurrence in Pingwu County, Sichuan Province, China. Among them, BoTNet and ConViT had the highest accuracy, both at 87.78%, an improvement of 1.11% compared to a single model, while ConViT had the highest F1-socre at 87.64%, an improvement of 1.28% compared to a single model. The results indicate that the fusion model of CNN and ViT has better LSM performance than the single model. Meanwhile, the evaluation results of this study can be used as one of the basic tools for landslide hazard risk quantification and disaster prevention in Pingwu County. [ABSTRACT FROM AUTHOR]

Landslides are serious natural disasters frequently occurring in the mountainous areas along the Sichuan–Tibet railway. Landslides are often induced by rainfall and greatly threaten people's lives and property. Via the use of GIS technology, a probabilistic analysis model was proposed for rainfall-induced shallow landslides along the above route. The process of the proposed probabilistic analysis model could be divided into three parts: (i) an infiltration hydrogeological model, (ii) a slope stability model, and (iii) probabilistic analysis. In this model, soil cohesion and internal friction angle were regarded as random parameters. The Green–Ampt model was used to dynamically describe the soil infiltration process during rainfall. Combined with the infinite slope stability model, the failure probability in the study area was analyzed. The model was applied to simulate the Bayi catchment landslide event of August 18, 2010, to evaluate its reliability. Through comparison of the simulation results to landslide occurrence locations, it was determined that the model achieves a satisfactory prediction performance. In addition, compared to deterministic analysis methods and the r.slope model, the proposed probabilistic analysis model achieved a satisfactory evaluation performance. The presented physics-based probabilistic analysis model could provide important theoretical support for hazard prevention in regard to rainfall-induced shallow landslides in mountainous areas. [ABSTRACT FROM AUTHOR]

Landslide is regarded as one of the most prevalent and destroying geological hazards in natural terrain areas. Reliable landslide susceptibility analysis procedures are vital for policymakers to manage the regional-scale landslide risk. In the framework of physically based modelling analysis, the infinite slope model is commonly used to assess the surficial landslide susceptibility with deterministically defined geotechnical and geological parameters. This work aims to develop a user-friendly geographic information system (GIS) extension tool called the GIS-FORM landslide prediction toolbox using the Python programming language to consider the possible uncertainties in the physically based landslide susceptibility analysis in seismic areas. We implement the first-order reliability method (FORM) algorithm to calculate the probability of infinite slope failures. The proposed toolbox can produce some regional hazard distribution maps of different indexes, such as the factor of safety (FoS), reliability index (RI), and failure probability (Pf). Furthermore, the toolbox enables coseismic landslide displacement prediction using either the direct Newmark integration method and/or the empirical formula method. Outputs of the GIS-FORM landslide prediction analysis are verified using published data in the literature. Further, it is also successfully employed for landslide susceptibility analysis of the Ms 7.0 Jiuzhaigou earthquake in Sichuan Province, China. Without loss of generality, the GIS-FORM landslide prediction toolbox can serve for the rapid hazard mapping of earthquake-induced regional landslides where uncertainties in geological and geotechnical parameters should be considered. [ABSTRACT FROM AUTHOR]

Rapid and accurate identification of landslides is an essential part of landslide hazard assessment, and in particular it is useful for land use planning, disaster prevention, and risk control. Recent alternatives to manual landslide mapping are moving in the direction of artificial intelligence—aided recognition of these surface processes. However, so far, the technological advancements have not produced robust automated mapping tools whose domain of validity holds in any area across the globe. For instance, capturing historical landslides in densely vegetated areas is still a challenge. This study proposed a deep learning method based on Light Detection and Ranging (LiDAR) data for automatic identification of historical landslides. Additionally, it tested this method in the Jiuzhaigou earthquake-hit region of Sichuan Province (China). Specifically, we generated a Red Relief Image Map (RRIM), which was obtained via high-precision airborne LiDAR data, and on the basis of this information we trained a Lightweight Attention U-Net (LAU-Net) to map a total of 1949 historical landslides. Overall, our model recognized the aforementioned landslides with high accuracy and relatively low computational costs. We compared multiple performance indexes across several deep learning routines and different data types. The results showed that the Multiple-Class based Semantic Image Segmentation (MIOU) and the F1_score of the LAU-Net and RRIM reached 82.29% and 87.45%, which represented the best performance among the methods we tested. [ABSTRACT FROM AUTHOR]

Understanding the spatiotemporal evolution and driving mechanisms of landslides following a mega-earthquake at the catchment scale can lead to improved landslide hazard assessment and reduced related risk. However, little effort has been made to undertake such research in the Wenchuan earthquake-affected region, outside Sichuan Province, China. In this study, we used the Goulinping valley in the Bailong River basin in southern Gansu Province, China, as an example. By examining the multitemporal inventory, we revealed various characteristics of the spatiotemporal evolution of landslides over the past 13 years (2007–2020). We evaluated the activity of landslides using multisource remote-sensing technology, analyzed the driving mechanisms of landslides, and further quantified the contribution of landslide evolution to debris flow in the catchment. Our results indicate that the number of landslides increased by nearly six times from 2007 to 2020, and the total volume of landslides approximately doubled. The evolution of landslides in the catchment can be divided into three stages: the earthquake driving stage (2008), the coupled driving stage of earthquake and rainfall (2008–2017), and the rainfall driving stage (2017–present). Landslides in the upstream limestone area were responsive to earthquakes, while the middle–lower loess–phyllite-dominated reaches were mainly controlled by rainfall. Thus, the current landslides in the upstream region remain stable, and those in the mid-downstream are vigorous. Small landslides and mid-downstream slope erosion can rapidly provide abundant debris flow and reduce its threshold, leading to an increase in the frequency and scale of debris flow. This study lays the foundation for studying landslide mechanisms in the Bailong River basin or similar regions. It also aids in engineering management and landslide risk mitigation under seismic activity and climate change conditions. [ABSTRACT FROM AUTHOR]

Landslides are recurrent geological phenomena on Earth that cause heavy casualties and property losses annually. In this study, we use the Vp-k stacking and nonlinear waveform inversion methods of high-frequency receiver functions extracted from local earthquakes, to sequentially invert Poisson's ratios and S-wave velocities of the Quaternary Xishancun landslide, which is composed of three segments, i.e., h1, h2, and h3 from bottom to top. Our results show that Poisson's ratio values are generally higher than 0.33 and that the S-wave velocities vary from 0.1 to 0.9 km s−1. High Poisson's ratios (>0.44) are mainly distributed in the juncture regions between different segments, as well as the western edge of h2. These zones show significant variation in landslide thickness and are potentially hazardous areas. Low velocities of 0.05–0.2 km s−1 with thicknesses of 10–30 m are widely observed in the lower layer of the landslide. The high Poisson's ratios and low-velocity layer may be related to water-rich materials in these areas. Our study suggests that the high-frequency receiver functions from local earthquakes can be used to delineate geotechnical structures, which is valuable for landslide stability analysis and hazard mitigation. [ABSTRACT FROM AUTHOR]

This work aims to characterize the earthquake-triggered landslides (EQTLs) during the 2017 Jiuzhaigou earthquake and to describe the geological and geomorphological control of landslide hazards. The Jiuzhaigou-EQTLs were studied via field-based investigation and remote sensing techniques during few weeks immediately after the mainshock occurred. The Jiuzhaigou earthquake produced an 11.4-km2 landslide area with 5633 landslides, dominated by shallow disrupted landslides, rock falls, and rockslides/rock avalanche. Statistical analyses show that the Jiuzhaigou-EQTLs had low mobility with a modal apparent friction angle of 40–44°. The Jiuzhaigou-EQTLs mostly occurred at elevations of 2700–3500 m and below 900 m relative relief to the main riverbed, where small-scale landslides were situated at the lower to middle parts of hillslope and large-scale landslides were situated close to the ridge crests. Jiuzhaigou-EQTLs clustered on river valley-side at the upstream of Baihe River (Jiuzhai Paradise and Ganhaizi village), Rize gully, and Danzhu gully on the tectonically active region at the southwest side of seismogenic fault F2. Landslides are prone to failure on the carbonate rocks with anaclinal hillslopes, following the cataclinal and cata-plagioclinal hillslopes. Most Jiuzhaigou-EQTLs preferentially occur on the below low-level break line (L2) in V-shaped valley-side and between first-level (L2) and high-level break line (L1) in U-shaped valley-side formed from long-term paleo-glacier and river incision. We found that landslide frequency is determined by a combination of the earthquake source with ground deformation, slope distribution, and rock strength due to weathering and dissolution along the fractures. Ground surface deformation, rock type (strength), geological structures, and valley-side slope breaks dictated the Jiuzhaigou-EQTL distribution pattern and hazards. This study elucidates the roles that local conditions play in influencing the occurrence, distribution, and frequency of slope failures and provides a better understanding of EQTL phenomenon during the Jiuzhaigou earthquake. [ABSTRACT FROM AUTHOR]

On 8 August 2017, Jiuzhaigou earthquake, magnitude 7.0, hit northern Sichuan, China. As the earthquake-stricken area is located in the mountainous region with forest and low residential density, the main damage is to vegetation and roads by earthquake-triggered landslides. In this study, the core area of Jiuzhaigou natural reserve, one of the highest seismic intensity zones, is selected. The landslides are extracted by examining vegetation changes from the preearthquake and postearthquake images using the Normalized Difference Vegetation Index (NDVI) and are verified by slope. As most road damage in the mountainous region could be attributed to the landslides nearby, the impacts of landslide on road are studied based on spatial analysis and are used to infer occluded road damage. Then, a knowledge-based method for postearthquake road detection and road capacity assessment from preearthquake road data and postearthquake high-resolution remote sensing imagery is proposed, as well as the quantitative road capacity assessment indicators to classify the road grades. This method is evaluated using the Beijing-2 (BJ-2) satellite images over the study area acquired on 28 April and 9 August. Compared with visual interpretation results, the extraction accuracy reached 90% for landslides and 85% for postearthquake roads, indicating that the approaches are effective and promising for quick response to devastating earthquake in similar circumstances. [ABSTRACT FROM AUTHOR]

A consequent slope, comprising weak planes in the same dip direction along the slope face, is the most dangerous type of rock slope. Herein, a consequent landslide that occurred on May 5, 2018, in Mabian County, Southwest China, was investigated. This landslide occurred in a region with a complex geological background and strong geological activity and is considered a typical consequent landslide. Field investigations and aerial photographic interpretations revealed that the rock mass slipped along the contact surface of the strongly and moderately weathered rock mass, triggered by continuous rainfall. A depth-integrated continuum method was used to study the dynamic process of the landslide. The applicability of the depth-integrated method was verified by comparing the moving boundary of the landslide extracted from a video with the simulation results. The results show that the velocity of the landslide first increased and then decreased from the top to the bottom of the landslide and reached a maximum of 21 m/s in the middle of the landslide. Based on this method, we predicted the potential danger zones and velocity distribution of unstable masses next to the Mabian landslide. On comparing the calculated results of a depth-integrated continuum method with the Scheidegger method, in the case of the Mabian landslide, a reduction factor λ was added to the Scheidegger formula to achieve a modified formula that determines the velocity of such consequent landslides relatively accurately. This work could provide reference for the study of landslide velocity and disaster prevention. [ABSTRACT FROM AUTHOR]

This paper aimed to measure the fractal dimension values of the spatial distribution of landslides in different geomorphic units, stratigraphic and basin zones based on the fractal theory and geological investigation, taking the landslides in Nanxi county, Yibin city, Sichuan province as a case study. The results demonstrate that the fractal dimension value of spatial distribution of landslide in Nanxi county is 1.1858. After classification and partition, the fractal dimension value of each small region is all less than the whole region. The fractal dimension values of different geomorphic types are sorted as platform deep hilly area, middle hilly area, low mountain area, shallow hilly area and valley flat area. The larger the fractal dimension value, the greater the contribution of the geomorphic unit to the development of landslide disasters will be. In the fractal dimension values of the stratigraphic types, they are Jurassic, Triassic, cretaceous and quaternary in order. In Jurassic strata, the bedding and oblique slopes of sand-mudstone interbeds with strata dip angle for more than 15° develop landslides the most. Fractal dimension values of stratigraphic types reveal the influence degree of stratum lithology and tectonic action on landslide development. Besides, The fractal dimension values of basins are sorted as Huangsha River, Longtan River, Meixi River, Bailichong River, Macaoxi River, Luqu River and Changxing River. With a maximum fractal dimension value, the Huangsha River basin has the widest distribution area and the largest river discharge in the study area. The riverbank slope is prone to landslide disaster because of the stronger flow erosion. This indicates the control of river hydrogeology on the landslides.Therefore, based on the calculation of fractal dimension values and the geological investigation of landslides in the study area, the results indicate that the fractal dimension value of spatial distribution of landslides is the comprehensive reflection of topography, stratigraphy, lithology, structure and hydrogeology, and it is the quantitative expression of regional landslides development complexity and susceptibility. Moreover, the fractal characteristics of the spatial distribution of landslides can be an important significance to division for the landslides-prone area. The calculation of fractal dimension could be an effective method to predict the susceptibility of geohazards in a regional scale. [ABSTRACT FROM AUTHOR]

Landslide susceptibility mapping (LSM) is a useful tool to estimate the probability of landslide occurrence, providing a scientific basis for natural hazards prevention, land use planning, and economic development in landslide-prone areas. To date, a large number of machine learning methods have been applied to LSM, and recently the advanced convolutional neural network (CNN) has been gradually adopted to enhance the prediction accuracy of LSM. The objective of this study is to introduce a CNN-based model in LSM and systematically compare its overall performance with the conventional machine learning models of random forest, logistic regression, and support vector machine. Herein, we selected Zhangzha Town in Sichuan Province, China, and Lantau Island in Hong Kong, China, as the study areas. Each landslide inventory and corresponding predisposing factors were stacked to form spatial datasets for LSM. The receiver operating characteristic analysis, area under the curve (AUC), and several statistical metrics, such as accuracy, root mean square error, Kappa coefficient, sensitivity, and specificity, were used to evaluate the performance of the models. Finally, the trained models were calculated, and the landslide susceptibility zones were mapped. Results suggest that both CNN and conventional machine learning-based models have a satisfactory performance. The CNN-based model exhibits an excellent prediction capability and achieves the highest performance but also significantly reduces the salt-of-pepper effect, which indicates its great potential for application to LSM. [ABSTRACT FROM AUTHOR]

On 12 May 2008 the magnitude 8.0 Wenchuan Earthquake erupted on the eastern fringe of the Tibetan Plateau in China, triggering many landslides onto towns and villages, leading to substantial loss of life and damage to property and the infrastructure. Local characteristic cloudy weather posed serious difficulties upon regional remote observation for disaster rescue in the mountainous region. To overcome weather limits to remote sensing, the study advanced one quick methodology to extract primary regional landslides with 8-day Moderate Resolution Imaging Spectroradiometer (MODIS) Normalized Difference Vegetation Index (NDVI) and terrain slope information. The threshold of temporal NDVI jump (value of 0.4) was introduced to discriminate landslide induced sharp NDVI decline from seasonal variations of vegetation cover, while slope information can be used to further confirm land cover change caused by a landslide. The study indicated that the Wenchuan Earthquake induced about 170 km2 of landslides in the southern half of the Longmenshan Fault Zone, most of which was formerly covered by forest or shrub. An ecological restoration of the areas affected by landslides will therefore be quite difficult. In addition, the result emphasizes that regions with slopes over 15° are exposed to high landslide risk, which should be carefully taken into account for settlement and transportation planning. [ABSTRACT FROM AUTHOR]

On 8 August 2017 an earthquake (MS7.0) occurred within Jiuzhaigou County, Northern Aba Prefecture, Sichuan Province, China, triggering 4834 landslides with an individual area greater than 7.8 m2 over a more than 400 km2 region. Instead of correlating geological and topographic factors with the coseismic landslide distribution pattern, this study has attempted to reveal the control from seismic landslide susceptibility mapping, which relies on the calculation of critical acceleration values using a simplified Newmark block model. We calculated the average critical acceleration for each cell of the gridded study area (1 km×1 km), which represented the seismic landslide susceptibility of the cell. An index of the potential landslide area generation rate was defined, i.e., the possible landsliding area in each grid cell. In combination with PGA (peak ground acceleration) distribution, we calculated such indexes for each cell to predict the possible landslide hazard under seismic ground shaking. Results show that seismic landslide susceptibility plays an important role in determining the coseismic landslide pattern. The places with high seismic landslide susceptibility tends to host many landslides. Additionally, the areas with high potential landslide area generation rates have high real landslide occurrence rates, consistent with dominant small-medium scale landslides by this earthquake. This approach can aid assessment of seismic landslide hazards at a preliminary stage. Additionally, it forms a foundation for further research, such as the rapid evaluation of post-earthquake landslides and identifying highly impacted areas to help decision makers prioritize disaster relief efforts. [ABSTRACT FROM AUTHOR]