Landslide monitoring and early warning systems are the most successful countermeasures to reduce fatalities and economic losses from landslide hazards. The traditional strategies such as GPS and extensometers are relatively expensive and difficult to be installed in steep, high mountains. In this study, a MEMS (Micro Electro Mechanical Systems) based multivariate wireless monitoring sensor unit was used to build a monitoring and early warning system for rainfall-triggered landslides, as one of the most practical and cost-effective countermeasures for hard-reached mountains. The multi-source wireless monitoring system and its well-developed equipment were tested in a landslide-prone slope to monitor the triggering of landslides and debris flows in the Wenchuan earthquake region, China. The variations of several state variables were observed, including the soil moisture content, soil matric suction, rainfall, inclination and ground vibration. The results of a slope stability analysis were benchmarked with the in situ measurements to identify the multivariate early warning parameters for rainfall-induced landslides. The proposed early warning system for the slope stability analysis can provide a more accurate prediction for rainfall-induced landslides and debris flows in earthquake hit regions.

The quality of sampling data critically influences landslide susceptibility prediction accuracy. Current studies commonly use a 1:1 ratio of landslide to non-landslide samples, failing to reflect natural geographical variability. This study develops a region-specific framework by integrating SHAP (SHapley Additive exPlanation) analysis with twelve landslide conditioning factors (LCFs) and three progressive sampling strategies, aiming to create adaptive non-landslide point selection criteria tailored to unique environmental and geological characteristics. The strategies include (1) multi-ratio random sampling (1:1 to 1:200), (2) susceptibility-based sampling adjustments derived from pre-susceptibility analysis, and (3) LCF-based correction using the NDVI threshold identified through SHAP analysis. Results show that LCF-based correction achieved the highest performance, while a 1:5 ratio proved optimal in random sampling, aligning with regional characteristics. This framework demonstrates the importance of region-specific sampling strategies in improving landslide susceptibility prediction.

The evolution of landslides is influenced by the complex interplay of internal geological factors and external triggering factors, resulting in nonlinear dynamic changes. Although deep learning methods have demonstrated advantages in predicting multivariate landslide displacement, their performance is often constrained by the challenges of extracting intricate features from extended time-series data. To address this challenge, we propose a novel displacement prediction model that integrates Variational Mode Decomposition (VMD), Self-Attention (SA), and Long Short-Term Memory (LSTM) networks. The model first employs VMD to decompose cumulative landslide displacement into trend, periodic, and stochastic components, followed by an assessment of the correlation between these components and the triggering factors using grey relational analysis. Subsequently, the self-attention mechanism is incorporated into the LSTM model to enhance its ability to capture complex dependencies. Finally, each displacement component is fed into the SA-LSTM model for separate predictions, which are then reconstructed to obtain the cumulative displacement prediction. Using the Zhonghai Village tunnel entrance (ZVTE) landslide as a case study, we validated the model with displacement data from GPS point 105 and made predictions for GPS point 104 to evaluate the model’s generalization capability. The results indicated that the RMSE and MAPE for SA-LSTM, LSTM, and TCN-LSTM at GPS point 105 were 0.3251 and 1.6785, 0.6248 and 2.9130, and 1.1777 and 5.5131, respectively. These findings demonstrate that SA-LSTM outperformed the other models in terms of complex feature extraction and accuracy. Furthermore, the RMSE and MAPE at GPS point 104 were 0.4232 and 1.0387, further corroborating the model’s strong extrapolation capability and its effectiveness in landslide monitoring.

The precise extraction of displacement time series for complex landslides poses significant challenges, and conventional landslide prediction models often overlook the deformation impacts of displacement triggers. To address this, we introduce a novel composite model tailored for predicting landslide displacement. This model employs Variational Mode Decomposition (VMD) to isolate each displacement component, with optimization achieved through the groupwise coupling algorithm. Subsequently, Grey correlation analysis (GRA) is applied to quantitatively assess the dynamic correlations between various triggering factors and landslide displacement. This analysis informs the construction of a feature set predicated on these correlation factors. Integrating the time-series VMD module into the standard Transformer architecture facilitates the prediction of landslide displacement. This integration allows for the extraction of critical time-evolution features associated with the displacement components. Ultimately, the predicted displacements are aggregated and reconstructed. We validate our model using the Bazimen landslide case study, analyzing displacement monitoring data from 1 January 2007, to 31 December 2012. The values of the root mean square error and the mean absolute percentage error were 1.86 and 4.85, respectively. This model offers a more nuanced understanding of the multifaceted causes and evolutionary dynamics underpinning landslide displacement and deformation, thereby markedly enhancing prediction accuracy.