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]

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]