What triggers an avalanche?

Returning to the story of Lawrence Jones, who twenty years ago was caught in an avalanche - here's what happened next...

Lawrence - Two Swedish guys who'd followed us up, taken a higher traverse than us, and as they stopped, those two triggered the thing off. They just saw this thing obviously crack at the top and then just gather momentum, gather and gather and gather and gather, and then just hit me like a thunderbolt.

As a younger man I played a lot of rugby, and there are moments - not very often - but there's a moment every now and again where you get caught up in a scrum or a ruck of some sort, where you are literally lifted off the ground and there's nothing you can do about it. And you may as well just stop fighting. And you realise, I'll save my energy 'til I'm unpeeled and I can get back out again. It was like that, but a thousand times more. I was sucked underneath all the snow and then the lights were out. And the thing I remember most about that is the eeriness, this terrifying silence.

Professor of geohazards Jim McElwaine and avalanche forecaster Mark Diggins join Chris Smith and Phil Sansom. Jim explains what it means when an avalanche is triggered...

Jim - Like Mike was saying, it was most likely it was a slab avalanche. So you've got hard, rigid snow and it's on top of a weak layer, and if you put extra loading on it, it can cause the weak layer to fail. And then the fracture can propagate extremely quickly. So this is what the trigger was, most likely; these Swedish people, they overstressed the weak layer and started it.

Chris - And Mark, would you concur with that? It's basically putting more mass on the slope above where the group of skiers involved with Lawrence were?

Mark - It may not be as clear as that. So it could be that when Lawrence went off his snowboard and was on his feet, he may have penetrated the snow. So if you imagine when you're traveling along a snow slope, you have a globe of effect underneath your feet. And if that globe coincides with a weak layer you'll get a collapse underneath that weak layer. That then can propagate, and it can propagate great distances, and that's relative to the density of the surface layer: so if it's harder, it'll propagate great distances; if it's not so hard, it won't travel as far. But it could well be that Lawrence, by coming off a snowboard, he could have initially weakened the slope. And then the final straw, if you like, was another pair of skiers on that slope could have made it release again.

Producer Amalia Thomas has an experiment to demonstrate, so if you’ve got your baking tray, flour, cornflour and salt, it’s time to follow along...

Amalia - So I put my baking tray flat on the table and on it I spread, as evenly as I can, course salt. On top of the salt, I put the cornflour; again, another thin layer of cornflour. So now I put the plain flour on top of the yellow cornflour. And I'm going to slowly tilt it until all my ingredients avalanche.

So this experiment is interesting for a couple of reasons: first of all because the combination of salt and flour avalanches at the same angle - between 35 and 38 degrees - as snow does in real mountains. And secondly, the way that the avalanche of salt and flour happened resembles the way snow avalanches break off the side of a mountain as well.

Jim explains how these layers of salt and flour and cornflour relate to the layers of snow mentioned earlier...

Jim - Well what you've got here is two sets of particles with different properties. So the salt particles are large and they're behaving like what we would call a dry granular medium. So that's like dry grains of sand. The only force between them is friction, there's no cohesion, so they're quite weakly bonded and they can easily move. Whereas the flour is much smaller so you get electrostatic effects and effects from the humidity that makes the particles stick together, so they have a lot more strength. You have this mixture of the different flowers, I think the cornflour makes it even stronger. So you've got a lot of strength in this slab-like layer on top of this weak layer. And then as you increase the slope angle, this simulates increasing the load - so this could be like a skier moving over the top or a fresh snow falling on top - and then the weak layer fails when the force of gravity down the slope overcomes the friction in between the salt particles.

Chris - Can the same science be used to explain any other natural phenomena? Or is this purely something related to avalanches?

Jim - Well that failure mechanism is purely related to avalanches, but they're very much like all sorts of other natural disasters: geophysical flows, like pyroclastic flows, debris flows, rockslides. And we use the same mathematical models really for all of these different types of events.

Chris - So do you have effectively a computer simulation that you can plug in all of the parameters of, say, snowfall, or an environment, and work out what the likely risks are and what the likely outcomes will be if things went wrong?

Jim - Well that's the aim of the work that I do, but we're not there yet. And I would say at the moment our models are not that predictive. So we can fix parameters to agree with past events, but we can't really say what's going to happen in the future.

Chris - Easy question to ask, impossible to answer: well why is it going wrong?

Jim - Well part of the problem is until recently there's been very little really good data about how these things flow and what's going on. Because normally you have a big powder cloud on top, so you couldn't really see what was going on. But now we have all sorts of great data, a lot of it from Switzerland where we artificially trigger avalanches. So this can really give us insight as to what is happening and help us develop better models.

Chris - Can you actually put things into the snow so that when an avalanche does happen, they're carried along and you can monitor what happens?

Jim - That's a great idea. And the avalanche people in Austria are working very hard on that. The big problem is then how to collect these things afterwards and get the data back out of them. Because the radio waves and things don't travel very well through solid snow, so you can't really get the data out easily while the avalanche is happening. And it's how to find these things afterwards is the tricky problem at the moment.