Poor snow person
This video comes from a just-published paper studying the dynamics of avalanches using detailed computer simulations of the physics of the snow. Turns out that, according to the authors, they started with code that was written by Disney for use in simulating snow for the movie Frozen, so this is science with an assist from Disney.
The code wasn’t written to analyze the science, just to make the snow look good on screen. The code simulated the snow as single elements that could move and interact independently, and included some basic details like gravity to make it look realistic. To make it applicable to science, researchers led by a scientist at the Swiss Federal Institute of Technology started building in other properties of snowpacks, such as the porosity of different layers and how the particles of snow bind together.
It is well established that many large avalanches start when there is a weak layer in the snow. Early season snow often falls on warm ground and partially melts before it is chilled and buried later in the season. This process creates an unstable situation where there is a porous, weak layer at depth buried beneath large amounts of snow. Scientists have simulated how that weak layer can collapse and trigger an avalanche before, but they haven’t been able to get the mechanisms to work right. In other words, they can simulate that an avalanche happens, but they couldn't make reasonable models for how big it would be.
The newly built code takes into account more of the interactions between the ice particles and the air – how they deform when they respond to stress, how the deformation mechanisms change when motion begins, etc. The end result is simulations like this one, starting with the small snow person trigger and ending with a full avalanche simulation. Using this model, scientists could go to real areas where avalanches are possible, collect data on the snowpack and the topography, and begin to actually assess the sizes of avalanches that could occur as the season develops.
-JBB
Video credit and original paper: https://www.nature.com/articles/s41467-018-05181-w