Keywords

Introduction

Understanding of landslide mechanism and causes are required to establish an effective mitigation and remediation. Besides in situ measurement and monitoring of hydrological condition leading to landslide movement, numerical and laboratory modelling have been a common scientific practice to study and evaluate landslide causes and mechanisms. Results of research and studies related to the advancement of landslide modelling are available in recent literature including journal articles, conference proceedings, and research reports. The objective of the session 2.3: Landslide Modeling: landslide mechanics and simulation models in the World Landslide Forum 4 (WLF4) was to provide a recent developments and research studies pertinent to physical and numerical modeling related to landslides causes, mechanism and remediation techniques.

Twenty contributions from ten different countries (Austria, Australia, China, Crotia, Italy, Japan, New Zealand, Norway, United Kingdom, USA), after a review process, were finally accepted for publishing in this session. These papers can be divided in the following general topics:

  • Numerical modelling of landslide mechanism

  • Physical/laboratory modeling of landslide causes

  • Numerical modelling of landslide remediation.

Numerical Modeling of Landslide Causes and Mechanisms

Debris flow is very common to occur during heavy rainfall or melting of ice and snow in a steep terrain. Mitigation efforts of such a landslide requires a better model of flow behavior. Chalks et al. (2017) evaluate the application of Smooth Particle Hydrodynamic (SPH) model and Voellmy’s rheological model predict the run out behavior of small scale debris flow experiments with physically viable rheological parameters. Meanwhile, De Finis et al. (2017) uses a pseudo-3D model RAMMS DEBRIS FLOW to reproduce the debris flow dynamic in the Gadria (Italy) anomalous system. Mitigation of flow landslide requires a good knowledge of impact force. Mergili et al. (2017) presents r.avaflow with a focus on the sensitivity of the model outcome to input conditions and material parameters. Coumo et al. (2017) uses the LS-RAPID model to simulate initiation, motion, expansion and propagation up to the stopping of the Montaguto earthflow. Prodan and Arbanas (2017) also use 3D stability analyses using LS-Rapid software for different weathering grades of flysch rock mass. The analysis demonstrates that raising of ground water level in the landslide body and weathering of the flysch rock mass on the sliding surface will have the main influence on possible landslide reactivation and further landslide development. Ceccato et al. (2017) studies numerically the impact process evaluating the effect of the material constitutive parameters and shape of the flow front. The results show that the front shape not only influences the peak pressure, but also the evolution of the impact force with time.

Similarly, predicting the velocity of slow-moving landslides also requires a better model. Hamasaki et al. (2017) proposes an effective simulation model based on motion equations using a mass system model incorporating viscous damping. They validated the velocity prediction model with the Kostanjek slow-moving landslide in Crotia and the Takino landslide in Japan.

In many countries where the soil thickness is very thin, shallow landslides often occur under a specific rainfall characteristics. To understand the mechanism of a shallow landslide occurred in the Province of Trento, North-Eastern Italy, Sanzeni et al. (2017) conducted a back analysis within a simplified two-dimensional conceptual framework based on the formation of a perched water table. Xue et al. (2017) conducted coupled deformation-seepage analyses and uncoupled seepage analyses to examine the influence of long duration rainfall on the stability and deformation characteristics of red-clay slopes in China. They found out that the increases in the pore pressure will cause variations in the unit weight of the slope material resulting in different amounts of deformations along the slope. Aiming at quantifying the relative importance of the hydraulic behavior of the soil-bedrock interface and of the hydraulic properties of the soil cover, Greco et al. (2017) performed a sensitivity analysis of the factor of safety to variations of hydraulic parameters for pyroclastic slope of Mount Cornito (Italy). The obtained results highlight that the equilibrium of the slope during rainfall infiltration is affected not only by the hydraulic characteristics of the soil cover, but a major role is played by the permeability of the soil-bedrock interface. Hoyer et al. (2017) used Electrical Resistivity Tomography (ERT) monitoring data to assess the hydraulic properties of the sub-surface soil using finite element modeling in combination with an optimization routine.

To select and implement an effective mitigation measure, the understanding of landslide propagation is necessary. Cuomo et al. (2017) employed the “GeoFlow_SPH” model to simulate the propagation stage, and a least squared weighted residual method for the inverse analysis to model the propagation stage of a landslide in Hong Kong. Modeling of a large-scale landslide in complex geological condition is quite a challenge for many researchers and engineers. Bar and Baczynski (2017) describes the most recent efforts undertaken in attempt to model and understand the mechanism of a highly ductile active-passive landslide on New York Ridge in the northern part of Ok Tedi Mine in Papua New Guinea. Using stress-strain modelling, Bolla and Paronuzzi (2017) to identify two types of bi-planar failure mechanism of rock slopes, depending on the different strength ratios between the unstable block and the delimiting discontinuities. Kalsnes et al. (2017) discusses the challenges related to tailings dam stability, with emphasis on geomechanical issues related to identification of potential failure mechanisms.

In the seismic prone areas, seismic slope stability analysis should consider the effect of ground motion. Passqua et al. (2017) presents results of 2D and 3D numerical modelling to assess the effect of irregular topography. They found out that the topographic contribution needs to be given careful consideration by way of 3D simulations.

Physical/Laboratory Modeling of Landslide Causes

The occurrence of rainfall-induced slope failures/landslide is controlled by many factors such as rainfall characteristics, slope geometry, soil properties etc. Tiwari et al. (2017) conducted a series of laboratory rainfall-induced slope failures to examine the influence of the variations in the slope density and the applied rainfall. They found out that increase in the initial void ratio and intensity of the rainfall resulted in an increase in the pore water pressure as well as the seepage velocity. In cold-high mountain areas, rock avalanches are induced by melting of ice and snow. To clarify the influencing mechanism of ice to enhance the propagation of rock ice avalanches, Yang et al. (2017) carried out a series of small scale flume tests under different rock and ice particle mass and size. The tests indicate that the mixing of ice and rock particles played an important role to enhance the mobility of rock-ice avalanches.

Modeling of Landslide Remediation

Bio-stabilization is also an interesting research topic on the aspect of landslide remediation. Dias et al. (2017a, b) provides a review of the mechanical effects of roots on shallow landslides, and of the forces applied by roots, in order to evaluate the safety factor of a reinforced slope. In order to understand the effect of root reinforcement, many methods for evaluation of root reinforcement in shallow landslides have been developed. Dias et al. (2017a, b) also provides a review of some methods to analyze root reinforcement and its use on assessing slope stability, and a discussion on the advantages and disadvantages of the methods, as well as their limitations.