A research team led by Professor Johan Gaume and postdoctoral researcher Lars Blatny has developed a model that realistically captures how snow, sand and other granular materials behave. The accompanying open-source software "Matter" enables realistic simulations - from avalanches to industrial processes.
How does snow compress in an avalanche? How does sand flow through a silo? And why do granular materials sometimes behave like a solid and sometimes like a liquid? Researchers from the Department of Civil, Environmental and Geomatic Engineering at ETH Zurich and the WSL Institute for Snow and Avalanche Research SLF have developed a new model that addresses these questions and have released the corresponding simulation software to the public.
The model was first introduced in their 2024 paper published in the Journal of Fluid Mechanics. It combines two previously separate approaches: the so-called µ(I)-rheology for flowing behaviour and Critical State Soil Mechanics for solid-like behaviour. This allows granular materials to be described across the full range from solid to liquid, especially relevant for materials like snow, which can be quite cohesive ("sticky") and can show large density variations, for example during an avalanche.
Based on this theoretical foundation, the researchers developed the software "Matter". It uses the Material Point Method (MPM), a numerical technique that can simulate large deformations and complex movements. With "Matter", users can run computer simulations - for example, of avalanches, landslides or industrial processes involving granular materials. The software is freely available, runs on any laptop and can handle complex terrain. It is already attracting users from institutes across Europe.
"We wanted a model that can handle the diverse behaviour of cohesive granular media, such as snow, and a software that makes it directly usable," Lars Blatny explains. "Matter is the result of combining theory of granular mechanics with state-of-the-art numerical schemes."
References
- Blatny, L., Gaume, J.
external page Matter v1: An open-source MPM solver for granular matter
Geoscientific Model Development (2025), doi:10.5194/gmd-18-9149-2025
- Blatny. L, Gray, J. M. N. T., Gaume, J.
external page A critical state µ(I)-rheology model for cohesive granular flows
Journal of Fluid Mechanics, 997, A67 (2024), doi: 10.1017/jfm.2024.643