The Daguangbao landslide (DGBL), triggered by the 2008 Wenchuan earthquake, is a rare instance of super-giant landslides globally. The post-earthquake evolution of the DGBL has garnered significant attention in recent years; however, its deformation patterns remain poorly characterized owing to the complex local topography. In this study, we present the first observations of the surface dynamics of DGBL by integrating satellite- and ground-based InSAR data complemented by kinematic interpretation using a LiDAR-derived Digital Surface Model (DSM). The results indicate that the maximum line-of-sight (LOS) displacement velocity obtained from satellite InSAR is approximately 80.9 mm/year between 1 January 2021, and 30 December 2023, with downslope displacement velocities ranging from −60.5 mm/year to 69.5 mm/year. Ground-based SAR (GB-SAR) enhances satellite observations by detecting localized apparent deformation at the rear edge of the landslide, with LOS displacement velocities reaching up to 1.5 mm/h. Our analysis suggests that steep and rugged terrain, combined with fragile and densely jointed lithology, are the primary factors contributing to the ongoing deformation of the landslide. The findings of this study demonstrate the effectiveness of combining satellite- and ground-based InSAR systems, highlighting their complementary role in interpreting complex landslide deformations.

Landslides are characterized by strong suddenness and a wide range of damage; accurate prediction of their susceptibility is an important prerequisite for regional risk prevention and control. To address the difficulties in acquiring landslide inventories in complex terrain areas and the insufficient interpretability of existing prediction models, this study proposes a landslide susceptibility assessment (LSA) framework that integrates automated sample detection and interpretability analysis. The proposed framework is applied to Moxi Town, a typical alpine valley area in Sichuan Province, China. A Mask R-CNN instance segmentation model was introduced to achieve automated detection of landslide samples, resulting in a high-quality inventory containing 923 landslides. Based on the spatial relationships between the landslide inventory and influencing factors, a convolutional neural network (CNN) landslide susceptibility assessment model incorporating Shapley Additive exPlanations (SHAP) interpretability analysis was constructed. The CNN model was further compared with random forest (RF) and extreme gradient boosting (XGBoost) machine learning models. The results show that the AUC value of the CNN model has increased by 4.3% and 3.2% compared with the RF and XGBoost models, respectively, and it significantly reduces the pretzel effect of landslide susceptibility mapping (LSM). The results validate the reliability of the proposed framework, which can provide technical support for landslide disaster prevention and monitoring.

Since the Wenchuan earthquake in 2008, Old Beichuan County-town has experienced significant subsidence due to the disruption of the geological environment and the concurrent increase in precipitation. The ongoing land surface deformation poses a threat to the preservation and utilization of the Beichuan National Earthquake Ruins Museum (BNERM), as well as to the safety of urban residents’ lives. However, the evolutionary characteristics of surface deformation in these areas remain largely unexplored. Here, we focused on the BNERM control zone and employed the small-baseline subset interferometric synthetic aperture radar (SBAS-InSAR) technique to accurately measure land surface deformation and its spatiotemporal changes. Subsequently, we integrated this data with land cover types and precipitation to investigate the driving factors of deformation. The results indicate a slight overall elevation increase in the study area from June 2015 to May 2023, with deformation rates varying between −35.2 mm/year and 22.9 mm/year. Additionally, four unstable slopes were identified within the BNERM control zone. Our analysis indicates that surface deformation in the study area is closely linked to changes in land cover types and precipitation, exhibiting a seasonal cumulative pattern, and active geological activity may also be a cause of deformation. This study provides invaluable insights into the surface deformation characteristics of the BNERM and can serve as a scientific foundation for the protection of earthquake ruins, risk assessment, early warning, and disaster prevention measures.

Landslides are among the most common and destructive geological hazards in complex mountainous regions. Developing accurate and interpretable susceptibility assessment models is important for regional geological hazard prevention and spatial safety management. To address the limitations of traditional machine learning models in processing tabular geoscience data with limited samples and complex nonlinear relationships, this study took Qingchuan County, Sichuan Province, as the study area and introduced the Tabular Prior-data Fitted Network (TabPFN) for landslide susceptibility assessment. Its performance was systematically compared with four commonly used models: Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF), and Extreme Gradient Boosting (XGBoost). An evaluation system was built using 12 conditioning factors, including elevation, lithology, and distance to roads (DTR). The SHapley Additive exPlanations (SHAP) method was further used to identify the contributions and effects of the main controlling factors. The results showed that TabPFN achieved the highest observed predictive performance among the evaluated models, with an AUC of 0.7834, showing higher observed performance than traditional models such as LR, SVM, RF, and XGBoost, indicating its strong ability to capture nonlinear relationships and good applicability under limited sample conditions. SHAP analysis further indicated that elevation, lithology, and distance to roads were the main controlling factors influencing the model predictions. This study demonstrates the applicability of TabPFN for landslide susceptibility assessment in complex mountainous regions and provides a new methodological reference for intelligent geological hazard prediction and refined regional risk management under limited data conditions.