EESS talk on "Unified modeling of snow and avalanche mechanics using the material point method"
Event details
| Date | 27.02.2018 |
| Hour | 12:15 › 13:15 |
| Speaker |
Dr Johan Gaume, Research & Teaching Associate, CRYOS, EPFL Short biography: Dr. Johan Gaume is a research and teaching associate in the CRYOS laboratory of EPFL and a scientist at the WSL Institute for Snow and Avalanche Research SLF in Davos. His research is focused on the numerical modeling of snow fracture and avalanche mechanics in order to improve avalanche forecasting. Last summer, he was Visiting Scholar in the Department of Mathematics of UCLA where he worked on a new snow avalanche model in the research group of Joseph Teran who contributed to the snow simulations in the Disney movie “Frozen”. |
| Location | |
| Category | Conferences - Seminars |
Abstract:
As a snow scientist, the realism of snow deformation in the Disney movie Frozen was a great source of wonder. Behind the magic scenes, UCLA / Disney mathematicians and computer scientists developed a snow model called "Matterhorn" based on the hybrid Eulerian-Lagrangian Material Point Method (MPM) with a simplified snow mechanical model. I was convinced that working together and combining our skills in mathematics and snow physics could improve the original model to simulate the complex fracture processes required for the release and flow of slab avalanches in view of ultimately improving avalanche forecasting and risk management.
Dangerous snow slab avalanches start with the failure of a weak snow layer buried below a cohesive snow slab. After failure, the very porous character of the weak layer induces its volumetric collapse and thus closing of crack faces. This complex process, generally referred to as anticrack, explains why avalanches can be remotely triggered from flat terrain. Based on Critical State Soil Mechanics, we developed the first constitutive model accounting for mixed-mode failure and weak layer collapse. In particular, our model is able to accurately reproduce the onset and propagation dynamics of anticracks observed in snow fracture experiments using particle tracking velocimetry.
Finally, we simulated the release and flow of slab avalanches at the slope scale triggered either artificially (bombing) or accidentally (remote triggering by a snowman).
As a snow scientist, the realism of snow deformation in the Disney movie Frozen was a great source of wonder. Behind the magic scenes, UCLA / Disney mathematicians and computer scientists developed a snow model called "Matterhorn" based on the hybrid Eulerian-Lagrangian Material Point Method (MPM) with a simplified snow mechanical model. I was convinced that working together and combining our skills in mathematics and snow physics could improve the original model to simulate the complex fracture processes required for the release and flow of slab avalanches in view of ultimately improving avalanche forecasting and risk management.
Dangerous snow slab avalanches start with the failure of a weak snow layer buried below a cohesive snow slab. After failure, the very porous character of the weak layer induces its volumetric collapse and thus closing of crack faces. This complex process, generally referred to as anticrack, explains why avalanches can be remotely triggered from flat terrain. Based on Critical State Soil Mechanics, we developed the first constitutive model accounting for mixed-mode failure and weak layer collapse. In particular, our model is able to accurately reproduce the onset and propagation dynamics of anticracks observed in snow fracture experiments using particle tracking velocimetry.
Finally, we simulated the release and flow of slab avalanches at the slope scale triggered either artificially (bombing) or accidentally (remote triggering by a snowman).
Practical information
- General public
- Free
- This event is internal
Organizer
- EESS - IIE
Contact
- Prof. Michael Lehning, CRYOS