Skip to main content
European Commission logo print header

Multiscale Modelling of Landslides and Debris Flows

Final Report Summary - MUMOLADE (Multiscale Modelling of Landslides and Debris Flows)

The Initial Training Network MUMOLADE (Multiscale Modelling of Landslides and Debris Flows) deals with the numerical and physical simulation of landslides and debris flows. Landslides and debris flows are serious geo-hazards common to countries with mountainous terrains, and are likely to be aggravated by the impending global climate change. The complex nature of landslides and debris flows can be described as multi-phase and multi-scale. The scientific objective of MUMOLADE is to achieve better understanding of the triggering conditions, failure mechanisms and runout dynamics, to improve the prediction of the deposition pattern and impact forces, and to provide the basis for rational design of stabilization measures, protection structures and early warning systems.

MUMOLADE connects 21 partners in a consortium which engages key stakeholders over several sectors: government authorities, engineers, contractors, manufacturers and software developers. The overarching aim of the MUMOLADE project is to provide high quality training to multidisciplinary young research fellows, and to develop their ability to navigate the challenging field of advanced computational and physical modeling of natural hazards. Our training program to improve scientific and networking skills of the MUMOLADE fellows included workshops and summer/winter schools, for which we also provided grants to external fellows in the same research field from other European research institutions. Training the next generation of researchers and the research conducted during this project are equally important in developing methods for code validation and model calibration to enable us to translate the numerical models into a widely adopted method providing reliable predictions and reasonable strategies to deal with these most dangerous geo-hazards.

To date, 14 Early Stage Research fellows (ERS) and 2 Experienced Research fellows (ER) have been working under the MUMOLADE project on one of the research topics within: Advanced constitutive models, Mechanics of multiphase media, Localized deformation analysis, Discrete Element Method (DEM), Advanced Finite Element Method (PFEM), Computational Fluid Dynamics (CFD), and Experimental investigations on laboratory and in-situ models.

During the course of the project, each fellow has produced several reports: A literature report detailing the state-of-the-art was delivered at the end of their first year in August 2013. During the second period of their projects (2014-2015) the fellows produced 24 technical reports updating on project milestones.

In total, 66 publications were produced by MUMOLADE fellows from the results of their research in the form of 10 conference posters, 37 papers in conference proceedings, 8 book chapters and 11 publications in peer reviewed journals (of which 10 were published in SCI journals, such as “Computers and Geotechnic”, “Computer Methods in Applied Mechanics and Engineering” and “Physics Review Letters”). Several fellows participated and presented their research results at the annual Engineering Mechanics Institute EMI 2015 Conference of the American Society of Civil Engineers at Stanford University in June 2015. All fellows participated in the Particles Congress (PARTICLES 2015) in Barcelona in September 2015 during which a special session for MUMOLADE fellows was held.

Three ESR fellows were awarded the PhD title within the project time frame in 2015. Four ESR fellows finished their PhD thesis by end of 2015 and are waiting for the defences. Six further ESR fellows will complete and defend their PhD thesis in 2016. Only one ESR fellow is not able to complete the PhD thesis within MUMOLADE because the initially recruited ESR fellow (ESR1) regrettably resigned for personal reasons after working at UNOTT for 27.75 months. For the remaining 8.25 months, a new fellow was recruited to continue the research work of the former fellow. However, the newly recruited fellow will require further funding to continue his PhD research beyond the remaining 8.25 months.

The research of the MUMOLADE fellows, using established as well as modified physical and numeric models and field and laboratory studies, contributed new theoretical and practical knowledge of landslides and debris flows. Here follows a brief outline of our projects:

- A new constitutive model describing the transformation of debris from a solid-like to a fluid-like state for the topic of “Hypoplastic constitutive model for unsaturated soil and debris materials”.
- An extended three dimensional micromechanical model based on the capillary theory for the topic of “Micro-macro relations for unsaturated materials and their integration in boundary value problems”.
- Implementation of the second order work criterion for the detection of the onset of slope failures in “Rainfall induced slope failures: Simulation and validation through case studies”. Three different variations are presented, which are applicable to variably saturated porous materials.
- Implementation and validation of an extension regarding viscoplasticity of an existing elastoplastic constitutive model for unsaturated soils in “FEM regularization for post localized bifurcation in fluid saturated media”.
- Development of a novel fluid-particle model to study the shearing behavior of granular soils under different saturation levels, ranging from dry material via capillary bridge regime to higher saturation levels with percolating clusters.
- Extension of the novel Modified Cam-clay model for root reinforcement to optimise for unsaturated soil conditions. Stability to rainfall of variously vegetated slopes is investigated in both 2D and 3D.
- In “Mixed Finite Elements formulations for the solution of localization problems in plasticity”, various problems of geotechnical interest are tackled to demonstrate the capabilities of the formulation. The results demonstrate a better performance of the Mixed Finite Element in all study cases compared to similar models.
- Proposition of a new method to simulate debris flow based on a conceptual separation of the debris material into a fluid phase and a granular phase in “Hybrid avalanches - Debris flow simulations with Discrete Elements and Lattice-Boltzmann”. The resulting DEM-LBM method is efficient and easily parallelizable, enabling the simulation of real systems. It is able to accurately simulate free-surface flows of particle-fluid mixtures typical of debris flow. The special reorganization of fluid and particles patterns typical of debris flow are modelled by the DEM-LBM without further ad hoc modelling. The method is tested through a series of analytical and numerical validations, as well as through comparison with commercial codes, proving its robustness in different situations and geometries.
- Investigation of a single-layer two-phase model for saturated flow and a two-layer two-phase model for unsaturated flow in “Simulation of debris flow with mixture theory and CFD”. The model reproduces the characteristic shape of some flow fronts. The resulting PDEs are solved using a high-resolution non-oscillatory central difference scheme with total variation diminishing property. Numerical solutions in terms of the effects of the drag force and fluid bed friction demonstrate the effectiveness of the proposed two-phase saturated model. For unsaturated flows, numerical solutions demonstrate that the present two-layer model can describe appropriately describe the transition process of a saturated grain-fluid mixture into an under-saturated state, and also capture the phenomenon of phase separation between the solid and fluid constituents associated with unsaturated flows.
- Creation of a general algorithm applicable to a variety of geotechnical problems based on the second generation Particle Finite Element Method (PFEM-2). The resulting method is flexible enough to solve both multi-fluid flows and Fluid-Structure Interaction (FSI) problems with minimal changes. All the tools developed during the research work, together with special customizations for this particular problem are used. The numerical experiments show that the physics of the problem is correctly captured, with good correlation against experimental results.
- Discrete Element Method (DEM) was used to model granular flow, rigid walls and flexible barriers in “Modelling the Mechanical Behaviour of Flexible Structures against Debris Flow”. It was found that the rigid wall is exposed to higher impact force. Using a mesh size as large as D90 of the flow is acceptable in terms of mass retaining capacity of the flow. In addition, not fixing the bottom cable of flexible barriers might lead to complete loss of its retaining capacity in extreme events.
- “Centrifuge modelling of rainfall-induced landslides in unsaturated soil slopes” experimentally evaluates the initiation conditions of rainfall-induced landslides. The results demonstrate the importance of the rainfall intensity during landslide initiation. During light and moderate rainfall events, slopes resisted deformation even after prolonged periods. However, during heavy rainfall events, infiltration of excess amounts of water saturated the unsaturated soil zone.
- In “Experimental modelling of debris flows and impact in geotechnical centrifuge”, excellent concordance is found in the analytical and numerical validation of the granular-front mechanisms, and the simulation of rapid granular flows in an augmented acceleration field. The work highlights the importance of simplifying complex processes into reasonable time and length scales when describing the interactions of a geophysical flow.
The MUMOLADE fellows are highly qualified to work in the challenging field of landslides and debris flows and will pose a valuable contribution to the long-term fight against potentially devastating geo-hazards. The scientific research results of the fellows are useful for research institutions and stakeholders, i.e. government authorities, engineers, contractors and manufacturers, and will eventually help to retain the competitiveness of the EU stakeholders to make reliable predictions and to devise mitigation measures against potentially devastating geo-hazards.

For contact information, news on upcoming events and further details on research results, please visit the project website at www.mumolade.com.