Technical information, news, research, and opinion on avalanches, snow safety, and winter backcountry travel.

Thursday, October 28, 2010

New From The Canadian Avalanche Centre

What we see and what we feelJonsi

The Canadian Avalanche Centre has published a great guide for advanced recreational skiers:

Avaluator 2.0 is also available:

Trip planning form:

These are great products from a wonderful organisation.

Friday, October 22, 2010

Union Creek

I keep dreamin', you'll be with me and you'll never goNickelback

This post is in memory of Kevin Carter, Devlin Williams, and Phillip Hollins.

February 2008
Almost three years ago I sat in a Seattle coffee shop and read an article in The Stranger. The sharp words pricked a nice hole in the warm bubble of what was an otherwise quiet afternoon: three snowboarders disappeared in Union Creek during early December, and they had not been seen, nor heard from since.

It was a La Nina winter, and you can bet snow was on my mind: I'd already visited friends in Revelstoke once during early January, and my upcoming travel plans included another trip to Revelstoke, followed by a trip to Banff to see my aunt and uncle.

So, I won't really get into what was going on in my life at that time, but I think massive changes sums it up quite nicely. In the past three months I had finally started to escape from a very dark hole, and I wanted nothing more than to stay in British Columbia and spend the rest of the winter climbing my way out of the awful black.

Not yet ready to head back to the States, I extended my trip and spent some time skiing near Valhalla Provincial Park. Sitting alone in my hotel room one evening, I realised that computer graphics techniques already provided solutions to the certain types of perception problems. We look at maps, and we get an idea about where we should and should not travel in general. But the computer can transform a map from merely useful to truly useful.

Theory of Relativity
In addition to incredible ease of access, the backcountry near Crystal Mountain is middle ground terrain. This sets up a classic middle ground perception problem because Union Creek has a rather benign appearance relative to high alpine terrain. In other words, you can drive to the trailhead, skin from the car, but you won't see enormous snowfields below savage peaks. Instead, as mountain terrain goes, Union Peak is really sort of small, steepish, and extensively gladed. To this point, there are plenty of places that seem safe..

In a typical scenario, the skier examines the choices in front of them and selects what 'appears' to be a safer option. A common equation is as follows: trading steepness for trees, or trading open slopes for tree covered slopes. However, safety is relative, and safer is not the same thing as safe. When avalanche danger is High, statistically speaking, the safer option might not actually be any safer at all. Without hard numbers, how do you know if you're really reducing or exposure or if you're just trading horses?

Can you identify ski runs that are safe during high avalanche danger?

You can examine the terrain from the air, and can you stare at contour maps all day long, but the fact of the matter is that the human mind just isn't very good at certain types of tasks. Computers are wonderful tools, and they are quite happy to help us cut through the clutter of our minds and tell us the truth.

Want to take a look at what the computer sees?

Union Creek
Union Creek is popular backcountry terrain east of Crystal Mountain Ski Resort. Union Creek contains COMPLEX avalanche terrain. In poor conditions, line-of-sight is limited and safe travel may be difficult for anyone regardless of skill level. Start zones in Union Creek are capable of producing large, destructive avalanches.

Required Skills: Expert route finding and expert snow stability assessment.

Visualization of terrain > 25 degrees and < 25 degrees. Avalanches often run into blue terrain.

Avalanche Terrain!

Terrain Trap!

Elements of avalanche terrain at Union Creek.

HistoryPeople have been killed by avalanches in Union Creek. Large soft-slab avalanches run during and after storms. Surface hoar formation is widespread in this drainage. Faceted snow often develops near ridgeline and rocks. Wind-loading occurs during high winds. Local skiers report 1, 2, and 3 foot crowns. ( Crown size of 30cm to 1m. )
Avalanche Starting ZonesMultiple avalanche starting zones are found below ridgelines.
Avalanche PathsChanneled and unconfined avalanche tracks are found throughout Union Creek. Avalanches move very quickly in confined tracks. On the east and north slopes, several large, poorly defined paths exist below large, open start zones. Numerous, small avalanche paths run through trees on all the slopes of Union Creek. Avalanches in these narrow, tree-lined paths are extremely dangerous.
Avalanche Runout ZonesA network of gullies forms overlapping runout zones at the bottom of Union Creek. This region is a complex, dangerous terrain trap. Other runout zones extend into forested areas.
ShapeConvoluted, with very frequent changes between concave and convex slope shapes.
Large Surface AreaTerrain in this drainage has a very large surface relative to its size on a map. Large amounts of snow accumulate throughout Union Creek.
Steep TerrainMuch of the terrain in Union Creek is very steep and avalanche prone. In many areas, more than 90% of the terrain suitable for skiing is between 30-40 degrees.
Open TerrainUnion Creek has large areas of open terrain.
ConfinedTerrainSome locations in Union Creek feature highly enclosed terrain.
Line-Of-SightSome locations in Union Creek have limited line-of-sight. You may not be able to see overhead avalanche terrain because of terrain features or trees.
Terrain TrapsUnion Creek has numerous terrain traps such as convexities, trees, depressions, and gullies. Computer modeling finds hundreds of terrain traps.
Safe AreasRidge areas are safest. Terrain on the valley floor, if below open terrain above, is not safe. Many of the avalanche paths in Union Creek are enlarging every winter. There are safer areas in the valley below thick expanses of trees that run to the ridgeline.
Exposure Time
Many ski runs and travel routes are exposed to avalanches for their entire length and offer no chance to reduce exposure. Less dangerous routes do exist, but require expert route-finding relative to the current conditions. Deep snow and steep terrain often make uphill travel slow and difficult.

Tuesday, October 19, 2010

Dancing Around Uncertainty

But it's one missed step ... one slip before you know it—Sarah McLachlan

There is an interesting article on Friends of Berthoud Pass web site.

Our Friend Bob Berwyn at the Summit County Voice wrote recently about the growing need for basic avalanche awareness among “sidecountry” skiers in Colorado.

Avalanche awareness programmes are wonderful, but many of them do not present balanced thinking frames for beginner recreationists. Programmes that focus entirely on "avalanche awareness" do so at the expense of "avalanche uncertainty", which leaves students with only half the mental model that they need to make good decisions.

The winter snowpack is conditionally unstable, and very often this means that it can be difficult to determine the extent of instability and its parameters. Of course, this is often the point where human nature steps in and people find their own ways 'manage' the uncertainty.

Failing to proactively manage uncertainty is at the root of many avalanche accidents for the following reason: a highly uncertain mind is very susceptible to biases, speculation, rationalisation, and the disregarding of facts.

Since some degree of residual uncertainty always remains, and since the degree of uncertainty is often inversely proportional to the skill of the recreationist, it seems strange to focus so much effort on awareness without addressing the other side of the coin as well.

We've all seen this take place during online discussions, at avalanche awareness events and during level 1 classes. Someone asks a question and the instructor provides additional information along with a qualification or two. This leads to additional questions and qualifications, as the instructor and students dance around the uncertainty.

I believe avalanche awareness programmes and level 1 classes must explicitly address uncertainty and teach students that decisions should retain a conservative character when their uncertainty is high. This would present a much more balanced mental model to beginners, and it would also help smooth out the cognitive dissonance that arises when observations of the terrain, snowpack, and weather don't provide clear answers.

Further Reading

Wednesday, October 13, 2010

Q & A: La Nina

Look around, leaves are brown, There's a patch of snow on the ground... - Simon & Garfunkel

Several people have emailed me questions about La Nina in the last couple weeks. The general vibe: people want to know what a La Nina winter will mean for avalanche safety in Washington State.

From a meteorological perspective, a La Nina winter is colder and wetter, which means larger, more frequent storms, and lots of snow in the mountains. However, La Nina has no mysterious influence on the snowpack itself. Prolonged cold temperatures, which can occur during any winter, will either produce new snowpack instability or allow existing snowpack instability to linger for longer periods of time. Large storms, which can occur during any winter, always produce instability. Rising temperatures during storms always produces instability. Some or all of these events may happen more often in a La Nina winter, but the snow safety rulebook won't change. You can learn more details about La Nina years on Amar Andalkar's excellent web site.

What about avalanches during La Nina winters? The last La Nina winter was 2007-2008, and there were a record number of fatalities in Washington State. For example: there were 5 fatalities in Washington State during a single weekend in early December 2007. In North America, there were 14 fatalities during the month December. That's basically one fatality every other day.

So, where were we? ...Frequent storms and lots of snow in the mountains... Sounds perfect, right?

Yeah, it's perfect ... but like everything wonderful, there's a catch. According to The Avalanche Handbook, "direct loading by synoptic scale weather events causes most avalanches". In other words, lots of storms means lots of potential for skier-triggered avalanches. Avalanches that occur during storms are referred to as direct-action avalanches, and these avalanches occur during a storm or within 24-72 hours after the storm ends. It's a good bet that public avalanche bulletins in North America will often forecast High avalanche danger during the upcoming winter.

What's your strategy for touring on days when avalanche danger is High? Do you have a list of trips for different levels of avalanche danger? ( Feel free to post tips in the comment box. )

The Take-Away
  • There will be frequent storms.
  • Avalanche danger will be High during and after these storms.
  • Start thinking about terrain choices for the winter.
  • Always choose terrain appropriate for current conditions.
  • Develop a list of "go to" terrain that is appropriate for high avalanche danger.
  • Plan like your life depends on it. ( Because it might. )
  • Take an avalanche safety course if you haven't already done so. United States / Canada
  • Hire a guide for big trips.

Saturday, October 2, 2010

Mind Games

We're playing those mind games together, pushing the barrier-John Lennon

In my last post, I remarked on an avalanche described in Backcountry Magazine. As promised, here is some discussion that skips the armchair quarterbacking.

Conditions on the day of their ski tour were as follows: a snowpack with a history deep instability had been very recently loaded by rain that was followed by 18 inches of snow. Obviously, even a cursory evaluation of conditions on the day of the tour should have raised awareness about the presence of instability. That aside, if the potential for high instability wasn't obvious from the recent weather and the history of the snowpack, the results of an extended-column test performed by the skiers ( ECT 5 Q2 ) should have been conclusive.

Yet despite the history of the snowpack, and despite the recent loading, and despite snowpack tests, all of which pointed to widespread instability with low triggering energy, the group ventured toward avalanche terrain and triggered an avalanche large enough to bury, injure, or kill a person. Thankfully, no one was injured.

From reading the article, and from examining their terrain choices, it seems as if the skiers were confused and uncertain about instability. Yet the manner in which I have laid out the facts about conditions leaves absolutely no question about the presence and parameters of snowpack instability on the day of the avalanche. So, as with all forms of armchair quarterbacking, in hindsight the snowpack instability seems perfectly obvious...

Clearly something doesn't add up.

Armchair quarterbacking is linked to the old saw: hindsight is 20/20. Therefore, when analysing accidents, it's critically important to ask the right questions: Do we really know if the skiers collected information in the same manner as us? What I mean is, did the skiers systematically collect and integrate information about conditions on the day of the ski tour? The article makes it pretty clear that the skiers collected information, but we do not know if they did so systematically, and despite being 'experienced', they clearly did a very poor job of integrating the information into their forecast. Did the skiers even develop an avalanche forecast? We don't know, but it isn't mentioned in the article.

Unfortunately, 'experienced' backcountry skiers frequently make dangerous mistakes, and the reasons why 'experienced skiers' make serious mistakes is a subject of much discussion in the avalanche research community. The magazine article chalks the errors up to powder fever and human factors, but I think both of these are just red herrings. Uncovering the actual mistake is important if we want answers that are clear and useful. Powder fever is certainly dangerous, but repeating that message doesn't really help anyone when powder fever is not the underlying cause. ( In medicine, fever is a symptom that reflects an underlying condition. ) Careless oversimplification greatly reduces the value of discussing skier-triggered avalanches.

After reading the description of their ski tour and the avalanche, the fundamental mistakes became very clear: the ski tour was a somewhat organic, loose, and chaotic endeavour on a day when avalanche danger was high. The presence of chaos is an extremely important clue, because it reveals what really went wrong. Just think about it for a moment .... a chaotic ski tour ... poor choice of objective ... on a day when avalanche danger was high ...

In this case, hopefully we can agree that 'experienced' refers to a person who has engaged in several seasons of ski touring. Hopefully we can also agree that there is a relationship between skill and experience, but that being experienced is not the same as being skilled.

What would lead a party of  'experienced' backcountry skiers down the garden path?

Lack Of Planning: A Fundamental Error
At the beginning of this post I stated that snowpack instability was glaringly obvious when in hindsight we systematically collected and integrated information. Armchair quarterbacking is so easy precisely because it's very simple to systematically collect and integrate information after the fact. But if you're a backcountry skier, it's incredibly important to systematically collect and integrate information before, during, and after a ski tour.

Of course, the systematic collection and integration of information has important implications, doesn't it? And what might those be? Is it possible that the systematic collection and integration of information occurs only when a ski tour is properly planned and properly executed? What are the odds of properly executing a ski tour when the trip has been poorly planned? The article content makes it quite clear that the ski tour was chaotic. So, even if there was some degree of planning, the planning certainly wasn't done properly. Second, the ski tour was executed in a rather haphazard fashion.

We know that a variety of activities such as map reading, instability tests, and route-finding took place on the day of the ski tour. That these activities had taken place obviously lead the participants to believe that their ski tour was properly planned. Yet none of the skiers discuss a formal trip plan, and this leads me to conclude that their tour was not properly planned, but merely discussed instead. Poor planning allowed 'powder fever' and 'poor communication' to take control of the situation. It's very simple: if you don't take control of the situation, something will, and this is doubly true for situations where uncertainty is high. ( I've written about the effects of unmanaged uncertainty. )

The article describes the skiers as experienced. Yes, there's that ugly word again. Since when did experience serve as a substitute for proper planning and execution? Please allow me be perfectly clear: discussions are not planning and experience does not guarantee clean execution. Rather than chalking up another series of mistakes to 'powder fever', it is more useful to consider how and why the circumstances developed in the first place.

Planning is a reality check that you can't afford to miss. If you don't plan, you're being intrinsically careless. If you don't plan properly, you're not planning. If you plan and execute poorly, you fail. Most of the time, failing to plan and poor execution won't have any consequences.

Most of the time.

Failing to plan is a serious primary error that creates situations in which secondary errors will occur. Rather than focusing on becoming an experienced backcountry skier, choose to become a skilled backcountry skier instead.

A Formula For Winter 2010-2010:

Solve for PYST, where = P = Plan, Y = Your, S = Ski, T = Tours

Planning Tips
Here are some planning resources:
Planning Checklist
Here is a planning checklist:
  • What is the current danger level?
  • Who is going?
  • What are their phone numbers?
  • What is their contact info?
  • Is our objective appropriate for the current danger level?
  • Do we have powder fever?
  • Have I planned the trip?
  • Do I have a map?
  • Have I analysed the terrain beforehand?
  • Do I have a list of safe alternates?
  • Have I separated the trip into legs with times, distances, and elevations?

Friday, October 1, 2010

Ex Posto Facto

Interesting article in Backcountry Magazine. I'll spare you the armchair quarterbacking, but the pictures and accompany text tell a very interesting tale. There were several key mistakes made. I'll write a full post - sans quarterbacking - once I read the entire article.