THE SNOWPACK YOU CANNOT SEE

The Technique

1. A snowpack is a stack of weather events. Every storm, every wind event, every clear cold night and every sunny afternoon leaves a layer with its own character, and those layers sit on top of each other all season in the order they arrived. What matters is not any single layer in isolation but the relationship between them: a strong, cohesive layer sitting on a weak, poorly bonded one is the structure that produces slab avalanches. The slab is the thing that breaks away and slides. The weak layer is the thing that fails. Both are needed, plus a slope steep enough to move and a trigger, and the trigger is very often the rider.
2. Snow changes after it lands, and the direction of that change decides everything. Where temperatures through the snowpack are fairly even, grains tend to round off, pack together and bond, which builds strength over time. Where there is a strong temperature difference across a shallow depth, water vapour moves through the snow and grows angular, sugary grains that do not bond, which builds weakness. That second process is why cold, clear weather over a thin snowpack, especially early in the season, is bad news months later. It is also why the same amount of new snow can be harmless in one range and dangerous in another. The history matters more than the headline.
3. Learn the three weak layers you will hear about constantly. Surface hoar is the feathery crystal that grows on the snow surface on clear, cold, calm nights, beautiful and glittering and about as structurally sound as a house of cards. If wind or sun destroys it before the next storm, it is a non event. If it gets buried intact, especially in sheltered openings and shaded slopes, it becomes a persistent weak layer that stays reactive for a long time. Facets are the angular grains described above, formed by temperature gradients, and they behave like ball bearings. Crusts, formed by sun, rain or wind, are strong themselves, but facets frequently form directly above or below them, and a crust also gives a fracture a smooth, uninterrupted surface to run along.
4. Understand why persistent weak layers are the killers. Most avalanche fatalities involve a slide triggered by the victim or someone in their party, and persistent layers are the ones most likely to let that happen in ways that feel unfair. They heal slowly, sometimes over weeks or the whole season. They allow a fracture to propagate a long way, so a failure under one rider can pull down an entire bowl. They allow remote triggering, so you can be standing on flat ground or in the trees at the bottom and set off the slope above. And they are stubborn in a way that trains people to trust the slope: the same feature gets ridden repeatedly with no result, and then one party finds the shallow spot or adds the load that finally does it.
5. Take the obvious signs seriously, because they are the cheapest information you will ever get. Recent avalanche activity on similar aspects and elevations is the single most useful observation available to you. Whumpfing, the sound and feel of a weak layer collapsing under you, means the structure is failing right now and would run if the ground were steeper. Shooting cracks running out from your track mean the slab is cohesive and connected. Heavy new snow, strong wind moving snow, rain on snow and rapid warming are all loading events. Any of these should shift you toward simpler terrain immediately, without discussion, because they are direct evidence rather than inference.
6. Accept spatial variability as a hard limit on knowledge. Snow depth, layer thickness, wind effect, sun effect and the presence of a weak layer all change across a slope, sometimes over a few metres, for reasons buried under the surface: a rock, a bush, an old wind eddy, a slight change in aspect. This means that a test pit tells you about the snow in that pit. It does not tell you about the slope, and it certainly does not tell you about the next slope. The weak spot that triggers the whole feature is frequently a shallow, unremarkable patch that nobody would have chosen to dig in. When someone says they dug a pit and the slope was fine, what they have actually established is that one square metre of it behaved a certain way at one moment.
7. Know what snowpack tests can and cannot do. Tests such as compression tests and extended column tests, done properly on a representative slope, tell you whether a weak layer exists in that spot, roughly how much force it takes to fail it, and in the case of propagation focused tests, whether a fracture will spread once it starts. That is genuinely useful information, and a result that shows easy failure with clean propagation is a strong reason to change your plan. What tests cannot do is prove absence. A test that does not react is weak evidence, because you may simply be standing in a strong part of a variable slope. This is the asymmetry that matters: a bad test result should be able to shut your day down, but a good test result should never be allowed to open one up.
8. Use the professional forecast as your primary snowpack information and your own observations as a check on it. Forecasters have field teams, weather stations, avalanche reports from a whole region and months of tracked layers, which is a body of evidence no recreational party can assemble in a morning. Your job on the hill is not to independently rebuild that picture but to confirm or contradict it: does what you are seeing match the forecast, and if it does not, are conditions worse than described. If your observations are worse than the forecast, believe your observations and simplify. If your observations are better than the forecast, do not use that as a licence to upgrade your terrain, because you are sampling a tiny fraction of the range.
9. Let all this produce humility rather than paralysis. The honest position is that the snowpack is partly unknowable, that professionals with decades in the field say the same, and that this is precisely why terrain and travel habits carry so much of the load. You do not need to know exactly what is under the surface if you have chosen ground where being wrong is survivable. The riders who last a long time in this sport are not the ones with the best snowpack intuition. They are the ones who assume their intuition might be wrong and pick terrain accordingly.

Common Mistakes

Digging one pit and declaring the slope good. Fix: use pits to look for structure and to confirm forecast layers, and let only negative findings change your plan.

• Ignoring whumpfing because the slope you are on is not steep. Fix: treat a collapse as regional evidence that the layer is primed, and drop your terrain choice for the rest of the day.
• Assuming a settled, quiet week means a persistent layer has healed. Fix: track how long the layer has been in the forecast and look for explicit forecaster language about it becoming dormant, then still keep a margin.
• Sampling snow only where it is easy or convenient to dig. Fix: pay attention to shallow, rocky, wind scoured spots, because that is where a rider is most likely to reach a deep weak layer.

The Drill

On a low angle, low consequence slope with no overhead hazard, dig a full profile and identify every layer by hand hardness and grain type, then move thirty metres across the same slope and do it again. Standard: write down the differences between the two profiles before you look at your first set of notes. If you find no differences, you probably have not looked hard enough, so check hardness by feel layer by layer rather than by eye.

← Full survival & remote first aid curriculum