Using multi-source monitoring data to model and predict the displacement behavior of landslides is of great significance for the judgment and decision-making of future landslide risks. This research proposes a landslide displacement prediction model that combines Variational Mode Decomposition (VMD) and the Long and Short-Term Time-Series Network (LSTNet). The bootstrap algorithm is then used to estimate the Prediction Intervals (PIs) to quantify the uncertainty of the proposed model. First, the cumulative displacements are decomposed into trend displacement, periodic displacement, and random displacement using the VMD with the minimum sample entropy constraint. The feature factors are also decomposed into high-frequency components and low-frequency components. Second, this study uses an improved polynomial function fitting method combining the time window and threshold to predict trend displacement and uses feature factors obtained by grey relational analysis to train the LSTNet networks and predict periodic and random displacements. Finally, the predicted trend, periodic, and random displacement are summed to the predicted cumulative displacement, while the bootstrap algorithm is used to evaluate the PIs of the proposed model at different confidence levels. The proposed model was verified and evaluated by the case of the Baishuihe landslide in the Three Gorges reservoir area of China. The case results show that the proposed model has better point prediction accuracy than the three baseline models of LSSVR, BP, and LSTM, and the reliability and quality of the PIs constructed at 90%, 95%, and 99% confidence levels are also better than those of the baseline models.

The data collection in the automated monitoring of landslides is often characterized by large amounts of data, periodic fluctuations, many outliers, and different collection intervals. The traditional method of calculating velocity and acceleration using the differential algorithm for landslide acceleration relies on experience to select thresholds and produces a large number of false early warnings. A hybrid early warning method for the landslide acceleration process based on automated monitoring data is proposed to solve this problem. The method combines the conventional warning method, based on cumulative displacement, velocity, and acceleration, and the critical sliding warning method based on normalized tangent angle according to different strategies. On the one hand, the least-squares fitting of monitoring data inside a given time window is used to calculate various early warning parameters, improving data usage and lowering calculation error. On the other hand, a dynamic semi-quantitative and semi-empirical method is provided for the determination of the thresholds, which is more reliable than the purely empirical method. The validation experiments at the Lishanyuan landslide in southern China show that the hybrid method can accurately identify the accelerating deformation of the landslide and gives very few false warnings. The proposed method is practical and effective for systems that require automated monitoring and warnings for a large number of landslides.

Although electrical resistivity tomography (ERT) may gather the internal resistivity information from a landslide area in a large-scale, low-cost, and non-invasive manner compared to point-based sensor monitoring technology, the indirect resistivity information obtained cannot directly evaluate the landslide’s current mechanical status, such as stress, strength, etc. Based on ERT monitoring data, a framework for quantitatively and directly evaluating the evolution of the factor of safety (FOS) of landslides during rainfall is proposed. The framework first inverts ERT observation data using the inexact Gauss–Newton method based on multiple constraints to obtain a more realistic resistivity distribution, then calculates the saturation distribution using Archie’s equation, and finally calculates the FOS of landslides using the finite element strength reduction method. Twelve sets of numerical experiments were designed and carried out based on the synthetic data of a theoretical model. The experimental results show that the proposed framework is valid and reliable under various arrays, apparent resistivity noise, and uncertainty in the water-electric correlation curve, with the Dipole-Dipole array outperforming the others in terms of accuracy, sensitivity, and anti-noise capability. The proposed framework is significant in improving ERT monitoring and early warning capabilities for rainfall-induced landslides.

In the study of the seepage characteristics of layered rock slope under rainfall conditions, the majority of previous research has considered the hydraulic conduction to be isotropic, or only considered the anisotropy ratio of the hydraulic conductivity, ignoring the anisotropy angle. In the current study, a layered rock slope in the Pulang region was selected as an example. Then, based on the fitting parameters of the Van Genuchten model, pore water pressure sensitivity analyses of the layered rock slope were carried out. The anisotropy ratio and anisotropy angle were used to analyze the sensitivity of the seepage and stability of the layered rock slopes. The results show that as the anisotropy angle of hydraulic conductivity of layered rock slope decreased, the maximum volume water content of surface (MWCS) of layered rock slope gradually increased. Additionally, as the anisotropy ratio decreased and the anisotropy angle increased, the rising heights of the groundwater (RHG) of layered rock slope gradually increased. When the hydraulic conduction of layered rock slope was considered isotropic, the factor of safety (FS) tended to be overestimated. As the anisotropy ratio decreased and the anisotropy angle increased, the factor of safety (FS) of layered rock slope decreased. Prevention should be the objective for rock slopes with larger dip angles in the bedding plane in the Pulang region. This study provides feasible schemes for the evaluation of the seepage and stability of layered rock slopes in Pulang region of southwestern China.

Landslide early warning is a key technology for effective landslide prevention and control. However, the traditional landslide early warning mainly makes decisions through thresholds, and if the thresholds are not selected properly, it will lead to missing alarms and false alarms frequently. To resolve this problem, this study proposes a landslide early warning algorithm based on a K-means-ResNet model. This method uses the K-means method to cluster the landslide deformation state, and then uses ResNet to classify the landslide rainfall and deformation data, so as to realize the threshold-free judgment and early warning of landslide deformation state. The model was applied to the Zhongma landslide, Guangxi Province, China, and the Shangmao landslide, Hunan Province, China, for validation and evaluation. The results showed that the accuracy, precision and recall of the proposed model can reach 0.975, 0.938, 0.863 and 0.993, 0.993, 0.925, respectively, for classifying the deformation states of the two landslides, and the classification results are better than those of the baseline model. Compared with the threshold-based early warning method, the proposed early warning method does not require artificial determination of threshold parameters and can effectively identify landslide deformation states, which can not only reduce false alarms and missing alarms but also improve the reliability of early warning.

It is the core prerequisite of landslide warning to mine short-term deformation patterns and extract disaster precursors from real-time and multi-source monitoring data. This study used the sliding window method and gray relation analysis to obtain features from multi-source, real-time monitoring data of the Lishanyuan landslide in Hunan Province, China. Then, the k-means algorithm with particle swarm optimization was used for clustering. Finally, the Apriori algorithm is used to mine strong association rules between the high-speed deformation process and rainfall features of this landslide to obtain short-term deformation patterns and precursors of the disaster. The data mining results show that the landslide has a high-speed deformation probability of more than 80% when rainfall occurs within 24 h and the cumulative rainfall is greater than 130.60 mm within 7 days. It is of great significance to extract the short-term deformation pattern of landslides by data mining technology to improve the accuracy and reliability of early warning.