Areas with vulnerable ecological environments often breed many geological disasters, especially landslides, which pose a severe threat to the safety of people’s lives and property in these areas. To aid in landslide prevention and mitigation, an approach combining the coefficient of determination method (CF) and a deep neural network (DNN) were proposed in this study for landslide susceptibility evaluation. The deep neural network can excavate the deep features of samples and improve the accuracy of the susceptibility model. In addition, the logistic regression model (LRM) and support vector machine (SVM) were selected to create landslide susceptibility maps for comparison, which also involved the coefficient of determination method (CF). Based on landslide remote sensing interpretation and field investigations, a spatial database of mudstone landslides in the Xining area was established. Eight different conditional factors, including the elevation, slope, slope aspect, undulation, curvature, watershed, distance from a fault, and distance from a road, in the study area were selected as the evaluation factors to evaluate the susceptibility. The results revealed that four factors (i.e., the ground elevation, curvature, distance from a fault, and distance from a road) had relatively significant influences on the landslide susceptibility in the study area. Finally, the confusion matrix was used to evaluate the accuracy of the results obtained using the three methods, and the optimal result was selected to evaluate the landslide susceptibility in the study area. It was found that the combined CF-DNN method was more suitable for evaluating the landslide susceptibility in this area. Landslide susceptibility zoning was conducted to divide the study area into four sensitivity levels: low (32.65%), medium (35.12%), high (22.44%), and extremely high (9.79%) susceptibility. The high-risk areas were primarily distributed in the high-elevation areas along the eastern edge of the Huangshui Basin.

Landslides are geohazards of major concern that can cause casualties and property damage. Short-term landslide displacement prediction is one of the most critical and challenging tasks in landslide deformation analysis, and is beneficial for future hazard mitigation. In this research, a novel short-term displacement prediction approach using spatial-temporal correlation and a gated recurrent unit (GRU) is proposed. The proposed approach is a unified framework that integrates time-series instant displacements collected from multiple monitoring points on a failing slope. First, a spatial-temporal correlation matrix, including the pairwise Pearson’s correlation coefficients, was studied based on the temporal instant displacement data. Then, the extracted spatial features were integrated into the time-series prediction model using GRU. This approach combines both spatial and temporal features simultaneously and provides enhanced prediction performance. In the last step, a comparative analysis against other benchmark algorithms is performed in two case studies including the conventional time-series modeling approach and the spatial-temporal modeling approach. The computational results show that the proposed model performs best in terms of performance evaluation metrics.