Many gentle dip translational rock slides have taken place in the Three Gorges Reservoir, China. In order to study the mechanism of these translational rock slides, the authors use the Anlesi landslide as a typical case study to investigate in detail. Field investigations show that the slip zones of the Anlesi landslide were formed from a white mudstone in Jurassic red strata. X-ray diffraction and infrared ray analysis showed that the main mineral components of the slip zones are montmorillonite, illite, feldspar and quartz. Laboratory tests indicate that the slip zone soils are silty clay, of medium-swelling potential, the shear strength decreasing significantly as the slip zone attracts water and saturates.The main factors contributing to the Anlesi landslide are recent tectonic activity, incompetent beds, and intensive rainfall. Recent tectonic activity had caused shear failure along the incompetent beds, and joints within the sandstone. With the effect of intensive rainfall, water percolates to the incompetent beds along tectonic fissures, resulting in swelling of the soil material and high groundwater pressures within fissures in the strata. As a consequence, the Anlesi slope is prone to slide along these incompetent beds.Flac3D software was used to simulate the mechanism of the Anlesi landslide considering the rheological properties of soil and rock. The simulation results demonstrate that the stress, displacement and failure area changes with simulated creep time. The maximum displacement in the X direction reaches 7.59 m after 200-year simulated creep. Therefore, the mechanism of the Anlesi landslide can be illustrated considering the rheological properties of Jurassic red strata. 相似文献
This work was to generate landslide susceptibility maps for the Three Gorges Reservoir (TGR) area, China by using different machine learning models. Three advanced machine learning methods, namely, gradient boosting decision tree (GBDT), random forest (RF) and information value (InV) models, were used, and the performances were assessed and compared. In total, 202 landslides were mapped by using a series of field surveys, aerial photographs, and reviews of historical and bibliographical data. Nine causative factors were then considered in landslide susceptibility map generation by using the GBDT, RF and InV models. All of the maps of the causative factors were resampled to a resolution of 28.5 m. Of the 486289 pixels in the area, 28526 pixels were landslide pixels, and 457763 pixels were non-landslide pixels. Finally, landslide susceptibility maps were generated by using the three machine learning models, and their performances were assessed through receiver operating characteristic (ROC) curves, the sensitivity, specificity, overall accuracy (OA) and kappa coefficient (KAPPA). The results showed that the GBDT, RF and InV models in overall produced reasonable accurate landslide susceptibility maps. Among these three methods, the GBDT method outperforms the other two machine learning methods, which can provide strong technical support for producing landslide susceptibility maps in TGR. 相似文献