Search

Landslides in mountainous areas act as an important control on morphological landscape evolution and represent a major natural hazard. The dynamic characteristics of a landslide directly relate to the distance it traveled and the scale of the resulting disaster. Based on extensive field investigations, we explored the effect of impact scraping on high-position landslides. During a rapid landslide, impact scraping amplifies the volume of the landslide and the size of the area affected by the landslide. Without acknowledging this effect, it is easy to underestimate the risk presented by a given potential landslide. In this study, we investigate the impact scraping of landslides that travel for significant distances both vertically (high-position) and horizontally (long-runout). There are four developmental stages of high-position, long-runout landslides: high-position shearing, gravitational acceleration, impact scraping, and debris deposition. Impact scraping amplifies the scale of the disaster by drastically increasing the volume of the landslide debris. After accounting for the effect of impact scraping, the total volume of the landslide exhibits a strong correlation with its travel distance. Additionally, the material properties of the erodible layer influence the landslide mobility. High-position and long-runout landslides have multiple scraping modes, including the embedding and excavation mode, the entrainment mode, the pushing and sliding mode, and the impact and splashing mode. In this study, we aim to provide insight that will improve the disaster modeling and risk assessment of high-position landslides, as well as to offer theoretical support for high-position and long-runout landslide dynamics research, disaster prevention and mitigation, and first responder rescue operation planning.

Human activities, including extensive land use practices, such as deforestation and intensive cultivation, may severely affect the landscape, and have caused important changes to the extent of natural forests during the last century in Southern Italy. Such changes had a strong influence on the frequency of occurrence of natural hazards, including landslides. Being one of the most significant control factors of slope movements, any variation in land cover pattern may determine changes in landslide distribution. The study area is the Rivo Basin which is located in Molise (Southern Apennines of Italy), a region severely affected by landslides. We prepared multi-temporal land cover and landslide inventory maps, aimed at developing different susceptibility maps to evaluate the effects of land cover changes in the predisposition to landslides. Based on the observed land cover trends in the study area, we simulated future scenarios of land cover in the attempt to assess potential future changes in landslide distribution and susceptibility. By investigating the relationship between the spatial pattern and distribution of past land cover settings and location factors (as elevation, slope, distance to settlements), we were able to calibrate a land cover change model to simulate future scenarios. The obtained results give important information both regarding the impact of past trends of land cover changes on landslide occurrence and possible future directions. They could be useful to provide insights toward a better land management for the study area, as well as for similar landslide-prone environments in Southern Italy, contributing to establish good practices for future landslide risk mitigation.

Xiluodu hydropower station is located on the Jinsha River, the upper course of the Yangtze River, and is the third largest hydropower station in the world. In May 2013, impoundment of the Xiluodu Reservoir started and preliminarily operation began, and the reservoir area was subject to frequent landslides and collapses. On 23rd July 2013, Huangping landslide, with a volume of 20 × 104 m3, slid into the Xiluodu Reservoir and generated an impulse water wave that killed 12 people on the opposite bank. The reservoir water level will fluctuate by 60 m each year, between the elevations of 540 and 600 m, so any impulse wave generated by a landslide in the reservoir is a major concern in the preliminary impoundment phase. Ganhaizi landslide is the biggest landslide in Xiluodu Reservoir, with a volume of about 7,800 × 104 m3. This landslide is only 14 km upstream from the Xiluodu Dam. In particular, the landslide deformation has obviously intensified after impounding. This paper analyzes the terrain and deformation features of Ganhaizi landslide, and we propose that sliding on the Ganhaizi slope may occur at the front part of the landslide. The front part of landslide has a volume of around 300 × 104 m3; the elevation of the barycenter of this part is 510 m. The sliding mass is thus largely underwater, and can pose a threat mainly in the form of surges. Based on the underwater landslide terrain in the mountain reservoir area, the authors have modified the seabed landslide-generated surge source model of Grilli and Watts (2005) to form an initial source model for an underwater landslide-generated surge. Using this model, the authors predicted possible surge disasters that can be caused by a Ganhaizi landslide at both the dead water level of 540 m and the normal water level of 600 m in Xiluodu Reservoir. The surge disaster might mainly strike the area within 5 km along the river channel both upstream and downstream of the landslide area, but generally poses no threat to Xiluodu Dam. This paper aims to provide technical support for the safety of Xiluodu Dam and the reservoir area. The newly established underwater landslide-generated surge initial source model for mountain reservoir areas may serve as reference for calculation of underwater landslide in other reservoir areas around the world.

After the Gongjiafang tsunami incident, the China Geological Survey recognized the extreme importance of subaerial landslide-generated tsunamis (SLGT) in the Three Gorges Reservoir, western China. The experimental models presented in the present paper were rooted in two common failure types of high, steep slopes in the Three Gorges Reservoir. One model represents a rigid rock plunging into the water and the other is a granular cluster sliding into the water. A wide range of effective parameters were considered and studied by performing 74 laboratory tests. The effects of bed slope angle, water depth, slide impact velocity, geometry (three-dimensional size of the rigid block and grain size of the granular cluster), and volume on impulse wave characteristics were examined. Slide kinematics and impulsive wave features, such as wave shape, amplitude, run-up, and run-down, were studied and compared. Experiments showed that the failure type of the rock mass plays a key role in SLGT and strongly influences air–fluid–solid interaction. Thus, to forecast SLGT rapidly, the zone of complicated air–solid–fluid interaction can be treated as a “black box” whose output is a water wave with definable characteristics, such as amplitude. Based on the experimental results, two dimensionless equations for the estimation of primary wave maximum amplitude were developed and successfully verified for the two failure types. The study provides basic information needed to forecast tsunamis generated by slopes with the potential to fail in other reservoirs around the world. These data can also be a very useful resource for theoretical analysis or numerical model validation.

In this paper, we conduct a numerical simulation analysis of the catastrophic 2009 Jiweishan landslide in Wulong, Chongqing, China, which was characterized by its rapid velocity and long runout. The study is based on detailed field geological survey data, and employs DANW runout analysis software to simulate the debris flow motion process, and accumulation characteristics of the landslide. The best match to the actual characteristics of the debris flow process was obtained by applying the friction rheological model in the source and accumulation areas, and the Voellmy model in the scraping area. The simulation results estimated the duration of the landslide to be approximately 101s with a maximum velocity of 39.2 m/s. The initial source volume, total accumulation volume, and runout distance obtained from an analysis of the DANW numerical simulation results were in good agreement with the corresponding characteristics of the actual event. The modeling scheme and parameters employed here are expected to help clarify the motion process of the Jiweishan landslide and improve the accuracy of future hazard assessments in areas with similar geological, topographical, and climatic features.