Search

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]

The accuracy of landslide displacement prediction can effectively prevent casualties and economic losses. To achieve accurate prediction of the Majiagou landslide displacement in the Three Gorges Reservoir (TGR), China, a hybrid machine learning prediction model considering the deformation hysteresis effect is proposed. The real-time deep displacement measurements were captured by using in-place inclinometers with Fiber Bragg grating (FBG) sensors. The time series method was adopted to divide the total displacement into a trend term and periodic term. Trend displacement was determined by the geological condition and predicted by the fitting method. Periodic displacement was controlled by external factors such as rainfall and fluctuation of reservoir water level. Before making the prediction, the grey correlation analysis was adopted to confirm that the fluctuation of the reservoir water level was the main influence factor. In view of the deficiency that current prediction methods could not quantitatively determine the lag time of landslide deformation and thus select the influencing factors empirically, the dynamic analysis of the correlation between periodic influence factors and periodic displacement was carried out in this paper, and the deformation lag time was identified to be 18 days by using set pair analysis (SPA) method. Finally, the optimal influence factors were selected and the prediction model of Majiagou landslide based on support vector machine optimized by particle swarm optimization (SPA-PSO-SVM) was established. Results showed that the root mean square error (RMSE) and the mean absolute percentage error (MAPE) of the proposed SPA-PSO-SVM prediction model are 0.28 and 12.8, respectively. Compared with the PSO-SVM model, the prediction accuracy of the proposed model had been improved significantly. The reliability and effectiveness of the SPA-PSO-SVM prediction model is verified and it has apparent advantages while predicting landslide displacement with deformation hysteresis effect involved. [ABSTRACT FROM AUTHOR]

The environmental factors of landslide susceptibility are generally uncorrelated or non-linearly correlated, resulting in the limited prediction performances of conventional machine learning methods for landslide susceptibility prediction (LSP). Deep learning methods can exploit low-level features and high-level representations of information from environmental factors. In this paper, a novel deep learning–based algorithm, the fully connected spare autoencoder (FC-SAE), is proposed for LSP. The FC-SAE consists of four steps: raw feature dropout in input layers, a sparse feature encoder in hidden layers, sparse feature extraction in output layers, and classification and prediction. The Sinan County of Guizhou Province in China, with a total of 23,195 landslide grid cells (306 recorded landslides) and 23,195 randomly selected non-landslide grid cells, was used as study case. The frequency ratio values of 27 environmental factors were taken as the input variables of FC-SAE. All 46,390 landslide and non-landslide grid cells were randomly divided into a training dataset (70%) and a test dataset (30%). By analyzing real landslide/non-landslide data, the performances of the FC-SAE and two other conventional machine learning methods, support vector machine (SVM) and back-propagation neural network (BPNN), were compared. The results show that the prediction rate and total accuracies of the FC-SAE are 0.854 and 85.2% which are higher than those of the SVM-only (0.827 and 81.56%) and BPNN (0.819 and 80.86%), respectively. In conclusion, the asymmetric and unsupervised FC-SAE can extract optimal non-linear features from environmental factors successfully, outperforms some conventional machine learning methods, and is promising for LSP. [ABSTRACT FROM AUTHOR]

In this paper, we propose a methodology for landslide susceptibility assessment at a regional scale in Yunnan, southwestern province of China. A landslide inventory map including 3,242 landslide points was prepared for the study area. Five factors recognized as correlated to landslide (namely, lithology, relative relief, tectonic fault density, rainfall, and road density) were analyzed and mapped in geographic information system. An index expressing the correlation between each factor and landslides [called class landslide susceptibility index (CLSI)] was proposed in the study. While analyzing landslide distribution in a large area, point aggregation might be expected. To quantify the uncertainty caused by aggregation, class landslide aggregation index was proposed. To account for the importance of each of the factors in the landslide susceptibility assessment, some weights were calculated by means of analytic hierarchy process. We propose a weighted class landslide susceptibility model (WCLSM), obtained by the combination of CLSI values of each factor with the correspondent weight. WCLSM performance in the study area was evaluated comparing the results obtained by first modeling all landslides and then by performing a time partition. The model was run including only landslides that occurred before 2009 and then validated with respect to landslides that occurred after 2009. The prediction-rate curve shows that the WCLSM model provides a good prediction for the study area. Of the study area, 21.4 % shows very high and high susceptibility and includes the 87.7 % of the number of landslides that occurred after 2009. [ABSTRACT FROM AUTHOR]

Sliding mass of landslides highly endangered the area along travel path, especially landslides with long travel distance. It is necessary to develop an effective prediction model for preliminarily evaluating landslide travel distance so as to improve disaster prevention and relocation. This paper collected 54 landslides with 347-4,170 m travel distance triggered by the 2008 Wenchuan earthquake to discuss the effectiveness of various influential factors on landslide travel distance and obtained an empirical model for its prediction. The results revealed that rock type, sliding source volume, and slope transition angle were the predominant factors on landslide travel distance. The validity of proposed model was verified by the satisfactory agreement between observations and predictions. Therefore, this model might be practically applicable in Wenchuan earthquake area and other similar geomorphological and geological regions. [ABSTRACT FROM AUTHOR]

As one of the major problems of geo-engineering, landslides often influence the safety of linear engineering projects that cross mountainous areas. Therefore, when selecting suitable routes for such projects, it is important to assess their susceptibility to landslides. In this paper, we used a natural gas pipeline in the northeast of the Yunnan-Guizhou Plateau of China as a case study to analyze landslide susceptibility. Based on engineering geological analogy, the analytical hierarchy process, and the least-squares method, a regional landslide susceptibility assessment model was developed and was programmed using GIS ArcEngine components under the Visual Studio.NET environment. The landslide susceptibility along the Zhong-Wu natural gas pipeline from Zhongxian County to Wuhan was assessed based on this model and classified into five levels: very safe, safe, moderate, susceptible, and very susceptible. The high accuracy and prediction capability of the model were confirmed by comparing the model results with past landslide data and performing a prediction test. The results indicated that the assessment model used in this study is reliable and can be used for landslide susceptibility assessment and route selection in other areas. [ABSTRACT FROM AUTHOR]

On September 5, 2022, an Ms6.8 earthquake struck Luding County, Sichuan Province, China. Through creating a coseismic landslide prediction model, we obtained the spatial distribution of the triggered geological hazards immediately after the earthquake. Through collecting all available multi-source optical remote sensing images of the earthquake-affected area via UAV and satellite platforms, the exact information of coseismic landslide was achieved by pattern recognition and visual inspection. According to the current results, the Luding earthquake triggered 5336 landslides with a total area of 28.53km2. The spatial distribution of the coseismic landslides is correlated statistically to various seismic, terrain, and geological factors, to evaluate their susceptibility at regional scale and to identify the most typical characteristics of these failures. The results reveal that the coseismic landslides mainly occurred on the sides of the Xianshuihe fault (within 1.2 km) and Dadu River (within 0.5 km) in striped patterns. They are concentrated in the regions with an elevation range of 1000–1800 m, a slope range of 25–55°, and lithologies of acid plutonic rocks, mixed sedimentary rocks, and siliciclastic sedimentary rocks. Besides, the coseismic landslides of the Luding earthquake are smaller in size and shallower than those triggered by the 2008 Wenchuan earthquake and the 2017 Jiuzhaigou earthquake. Rapidly achieving the spatial locations and distribution patterns of the coseismic landslides enables to provide effective support and guidance to emergency rescue, risk mitigation, and reconstruction planning. [ABSTRACT FROM AUTHOR]

Loess landslides pose significant physical and environmental threats to communities of the Heifangtai Terrace in Gansu Province, China. Estimating potentially affected areas using dynamic numerical simulations provides a foundation for risk assessment and management. However, current studies often use rheological input parameters derived from the back analysis of an individual historical landslide, which requires that the potential landslide is quite similar to the historical landslide to guarantee the accuracy of predictions. We collected data from 20 loess flowslides that occurred from 2015 to 2019 in the Heifangtai Terrace. A friction model embedded in MassFlow was used to simulate run-out behavior. Accumulation depths at multiple control points were used as observation information. A multi-objective optimization back analysis was used to calibrate optimal rheological parameters for each flowslide, yielding a region-specific database of 20 datasets for statistical analyses. Finally, two flowslides that occurred in 2019 were employed as potential landslides to conduct probabilistic run-out predictions based on the database. These predictions agreed well with the observed accumulation areas in both cases, thus validating our method. This database will be useful for future landslide run-out predictions in this area. [ABSTRACT FROM AUTHOR]

Fifty-five descending images from the ENVISAT satellite were processed using the small baseline subset (SBAS) method to derive the spatial and temporal ground deformation of the Bailong River Basin between 2003 and 2010. The basin is one of the most severely landslide- and debris flow-affected areas of China. As a result, 104 sites with high deformation areas were identified. Interferometric Synthetic Aperture Radar (InSAR) analysis was combined with landslide inventory data and field surveys, and anomalous areas were classified into three main types: landslide; debris; and subsidence. Displacement rates up to 35 mm/yr were evaluated away from the sensor along a line-of-sight (LOS) direction. The results gained should allow a more accurate prediction and monitoring of landslides, debris, and subsidence; further, they demonstrate the capability of the SBAS method to analyze any displacement effect and identify dangerous and uninhabitable areas in the basin. The small baseline subset method can thus contribute to the prediction and prevention of geohazards in the area. [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]

A large-scale obliquely inclined bedding rockslide, activated by a heavy rainstorm, occurred on July 8, 2020, at 7:05 (UTC + 8) in Shiban Village, Songtao Miao Autonomous County, Guizhou Province, China. The loss of life in this event was greatly reduced owing to the local warning system for rainstorm-induced geohazards. To understand the failure characteristics, triggering factors, the genetic mechanism of the landslide, the geomorphological features, geological characteristics, hydrological conditions, and rainfall characteristics were systematically studied by a synthetic approach including field investigations, satellite imagery, unmanned aerial vehicle (UAV) photography, laboratory tests, and rainfall data statistics. The results indicated that the interface between the soft and hard rock, the well-developed joints, and the free face in front of the slope constituted the boundaries of this landslide. The concave topography at the back and southern edge of the landslide, the bare ground, and the cataclastic structure of the rock mass provided favorable conditions for the collection or infiltration of rainwater. The concentrated rainstorm was the direct trigger for the landslide, which led to a rapid inflow and retention of rainfall in the landslide through favorable landform and geological conditions. The groundwater recharge that cannot be drained in time caused the mechanical deterioration of rock mass and induced a rapid increase in pore water pressure in the landslide. Moreover, the water level of the Ganlong River at the toe of the slope also rose rapidly, and the uplift pressure in front of the slope increased accordingly. Under the combined action of these adverse factors, the overall anti-sliding force of the slope was less than the sliding force, finally resulting in the landslide. Remarkably, the local warning system for rainstorm-induced geohazards successfully forecasted the landslide, but the shortcoming is that the forecast time in advance is short. Nevertheless, the prediction has significantly reduced human casualties and provided valuable experience for the prediction of this type of landslide. [ABSTRACT FROM AUTHOR]

More than 5000 landslides or potential landslides have been identified in the Three Gorges Reservoir (TGR) region since its impoundment in 2003. These hazards seriously threaten the continuation of reservoir operations and the safety of dams, waterways, and local residents. Understanding the deformation characteristics and kinematic evolution of these landslides can be helpful for assessing their stability and providing long-term landslide predictions. In this study, the deformation characteristics, influencing factors, sliding mechanism, and threshold reservoir level of the Muyubao landslide are revealed. Data from a professional monitoring network implemented in 2006 reveal that the deformation of this landslide is in a stage of creep characterized by overall movement, and the rate of creep is gradually slowing. The deformation mechanism is investigated by combining meteorological, hydrological, and reservoir level data with displacement measurements from surface cracks, manual and automatic GPS monitoring records, and field investigations and drilling surveys. The reservoir level is the main driving factor of deformation, whereas the filling-drawdown rate is not directly related to the deformation of the landslide. Furthermore, rainfall can promote deformation during periods with high reservoir levels. The deformation of the landslide follows a very distinctive pattern featuring large displacements after the reservoir level rises to a high level (from late October to early March of the following year) and a rapid transition to constant deformation as the reservoir level drops. The results indicate that the reservoir level of 172 m is the general threshold reservoir level of the Muyubao landslide. At present, the deformation rate of the Muyubao landslide is gradually decreasing, and the probability of large-scale sliding is low; however, the monitoring system must be strengthened and additional research must be performed on the deformation mechanism. [ABSTRACT FROM AUTHOR]

Currently, mathematical models such as the regression model, grey prediction model, and neural networks are commonly used to predict landslide displacement. In this study, we develop multi-data-driven models for prediction of landslide displacement. Multi-kernel learning and weighted learning methods are integrated with the grey prediction theory and three models are constructed to predict landslide displacement: the kernel-based grey model, multi-kernel grey model, and weighted multi-kernel grey model. All three models are simple to calculate and easy to program. The models were applied to a case study of the Baishuihe landslide in the Three Gorges reservoir, China. The influence of rainfall, reservoir water level, landslide rate, and historical displacement on the current displacement of the landslide were investigated, and multi-data-driven prediction models with input parameters (rainfall, reservoir water level, landslide rate, and historical displacement) and an output parameter (current displacement) were developed. The results show that the weighted multi-kernel grey model has better prediction accuracy than the kernel-based grey model and the multi-kernel grey model. These findings can be widely applied in practical landslide prediction. [ABSTRACT FROM AUTHOR]

Many models have been widely used in landslide displacement prediction. However, few studies have proposed quantitative prediction formulas. Thus, the variational mode decomposition (VMD) theory was applied to decompose the "step-like" displacement of landslides into trend displacement, periodic displacement, and random displacement. Then, a novel prediction model based on wavelet analysis (WA) and a back-propagation neural network (BPNN) optimized by the grey wolf optimizer (GWO) algorithm was proposed (the GWO-BP model) to obtain a prediction formula. In this model, a polynomial function was first used to predict the trend displacement. All the hidden periods of periodic displacement were calculated using the WA method, and a trigonometric function was applied to predict the periodic displacement. In addition, based on an analysis of the grey relational degree (GRD), the main triggering factors, which can affect the random displacement, were determined. Then, the mathematical connection between random displacement and triggering factors was obtained with the GWO-BP model. Finally, all the predicted values were superposed to achieve the prediction cumulative displacement based on the time series model. The Outang landslide in the Three Gorges Reservoir area, China, was taken as an example, and the displacement data of monitoring sites MJ01 and MJ02 from December 2010 to December 2016 were selected for analysis. The results indicated that the root mean square errors (RMSE) between the real displacement values and the prediction values obtained using the formula were 14.79 mm and 12.59 mm, respectively. The correlation coefficient R values were 0.99 and 0.93, respectively. This model can be used to obtain the landslide displacement formula and provide a solid basis for developing early warning systems for landslides. [ABSTRACT FROM AUTHOR]

Through the mathematical description of landslide evolution, landslide behavior can be simply studied by the corresponding displacement time sequence. Thus, to predict landslides, a new intelligent prediction method that combines the gray system method (GM) and an evolutionary neural network (ENN) is proposed. In this method, the original displacement is divided into two components: the displacement trend and deviation sequences, which are predicted by the GM and a new ENN combined by immunized continuous ant colony optimization and modified back propagation algorithm. This new method is applied to two typical landslides in the Three Gorges Reservoir of China to show that the forecasting precision of the new method is highly accurate in these cases. For comparison with other methods (GM and ENN), the forecasting performance of the new method is analyzed in depth. The results show that the generalization prediction precision of the new method is much higher than those of the GM and ENN. Finally, the influences of the extrapolation prediction time and the prediction step for the new method are comprehensively analyzed. The results show that the lower the extrapolation prediction time and the prediction step are, the higher the extrapolation prediction precision. [ABSTRACT FROM AUTHOR]

Landslides often occur within the reservoir area behind dams. In China, a common strategy for stabilizing these landslides is to install large piles through the landslide and into the stable ground below. The piles interact with the landslide and constitute a landslide-stabilizing pile system. The deformation of this system under the reservoir operation is more complicated than the deformation of the landslide itself. Understanding the behaviour of this system is very important to the long-term safety of landslides stabilized with piles in reservoirs. The Majiagou landslide, which was selected as a case study, was triggered by the first impoundment of the reservoir behind the Three Gorges dam. A row of anti-slide piles was installed in the landslide in 2007, but monitoring results found these were ineffective at stabilizing the landslide. Subsequently, in 2011, two longer test piles and an integrated monitoring system were installed in the landslide to better understand the failure mode of the landslide and to measure the deformation characteristics of the landslide-stabilizing pile system. Monitoring results show that the Majiagou landslide is a translational landslide with three slip surfaces. The test piles provided local resistance and partially slowed down the sliding mass behind the piles, and the landslide deformation response to external factors decreased for a time. However, after 2 years, the deformation of the landslide-stabilizing pile system reverted to seasonal stepwise cumulative displacements influenced by cycles of reservoir drawdown and rainfall. The monitoring results provide fundamental data for evaluating the long-term performance of anti-slide piles and for assessing long-term stability of the stabilized landslide under the reservoir operation. [ABSTRACT FROM AUTHOR]

Landslide prediction is important for mitigating geohazards but is very challenging. In landslide evolution, displacement depends on the local geological conditions and variations in the controlling factors. Such factors have led to the “step-like” deformation of landslides in the Three Gorges Reservoir area of China. Based on displacement monitoring data and the deformation characteristics of the Baishuihe Landslide, an additive time series model was established for landslide displacement prediction. In the model, cumulative displacement was divided into three parts: trend, periodic, and random terms. These terms reflect internal factors (geological environmental, gravity, etc.), external factors (rainfall, reservoir water level, etc.), and random factors (uncertainties). After statistically analyzing the displacement data, a cubic polynomial model was proposed to predict the trend term of displacement. Then, multiple algorithms were used to determine the optimal support vector regression (SVR) model and train and predict the periodic term. The results showed that the landslide displacement values predicted based on data time series and the genetic algorithm (GA-SVR) model are better than those based on grid search (GS-SVR) and particle swarm optimization (PSO-SVR) models. Finally, the random term was accurately predicted by GA-SVR. Therefore, the coupled model based on temporal data series and GA-SVR can be used to predict landslide displacement. Additionally, the GA-SVR model has broad application potential in the prediction of landslide displacement with “step-like” behavior. [ABSTRACT FROM AUTHOR]