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

Sunday, March 28, 2010

Details, Details, Details

I've seen clouds from both sides now—Joni Mitchell

This is a line from one of my favourite songs. Anyway, I'm not sure if this song is so good because the lyrics are so universally brilliant, or just because Joni is a goddess. The message of the song, just when you think you've got it all figured out..., is very relevant for avalanche safety.

With that in mind, what is wrong with the following picture?

Wednesday, March 24, 2010

Exposure Is Personal

We reached the dizzy heights of that dreamed of world—Pink Floyd

NOTE: I updated this post on March 28th, 2010. I don't include the role of avalanche size in the discussion on danger because it is possible to be killed by a very small avalanche.

Often times you hear people discuss "mitigating risk", or "mitigating hazard", and so forth. From these discussions, it's clear that many people can't distinguish between hazard, danger, and risk. ( Which is perfectly understandable since these words are fuzzy and woefully imprecise. )

There is a difference between exposure, danger and risk, and in addition to the differences, it's also very important to understand how these variables interact. Risk is classically defined as chance, consequences, and exposure, but in this post I'm going to explicitly re-fashion the key elements of risk to revolve around your personal risk acceptance level, since the real source of most risk is from within.


You are always exposed to avalanches during travel through avalanche terrain—that's why it's called avalanche terrain. Usually during times of high avalanche danger, you can reduce exposure by choosing gentle terrain, or by using very careful route-finding / travel technique when traveling through exposed terrain.


The actual avalanche danger may be very low or very high. The danger level reflects the likelihood of avalanche formation, as outlined in the public avalanche bulletin, or as judged by your own observations in the backcountry. You cannot modify the current avalanche danger, but through informed terrain choices, you may be able to choose slopes where avalanche danger is lower.


Your personal risk acceptance level determines your tolerance for exposure and danger, which in turn determines the chance of avalanche involvement and the consequences thereof. This is why avalanche professionals often say most avalanche accidents are the result of choice, not chance. Again, the most important source of risk comes from within.

Scales For Exposure, Danger, And Risk

Figure 1.1. Avalanche danger scale from LOW to EXTREME. This graph shows the likelihood of avalanche formation, and this information is available in the public avalanche bulletin, or formed by your own observations in the field.

Figure 1.2. Terrain Exposure Scale. This graph shows the amount of exposure for a given route or geographic area. One problem is that there are few terrain ratings available for Washington State. ( I've assembled ratings for some popular areas, but the list is still very incomplete. )

Figure 1.3. Risk Acceptance Scale. This graph shows the amount of risk you are willing to tolerate. Baseline risk tolerance varies greatly by individual, and specific individuals have different risk tolerances at different times as well. You may generally accept higher risk, but even this will vary.

Your Personal Exposure Score

Calculate your personal exposure by using the following formula.

risk tolerance + avalanche danger + avalanche exposure = personal exposure score
  • Avalanche Danger Score: Current danger level, 1-5, multipled by 2.
    • Considerable = 6.
  • Terrain Exposure Score: A number from 0 to 10.
    • 0 = no exposure.
  • Personal Risk Score: A number from 0 to 10.
    • 0 = low risk tolerance.
One of the most important, and least discussed elements of avalanche safety, is the complicated relationship between exposure, danger, and risk, and the process of managing awareness and uncertainty. Reducing exposure commonly makes people feel less uncertain, even though a reduction in exposure may not have the desired effect when you actually want more risk.

Always evaluate the moments when you seek to preserve risk. Often these are moments when we look for something fun, exciting, or notable.

Further Reading

Thursday, March 18, 2010

The 7 Deadly Sins

if I could turn back time—Cher

This list is composed mostly of my own thoughts about what constitutes the seven deadly sins of winter backcountry travel. Your mistakes may differ, and your mileage may vary.
  • Lack of preparedness.
  • Failure to maintain awareness.
  • Failure to manage uncertainty.
  • Poor communication.
  • Not paying attention.
  • Ignoring margin-of-safety.
  • Failing to reduce expectations.
Feel free to suggest additional mistakes.

Wednesday, March 17, 2010

Behavioural Strategy

I can't live, with or without you—U2

I've been busy with job, family, and snow. So I haven't really had a chance to properly finish two posts on which I am currently working. My day job involves business and software, so for the business side of my life, I subscribe to the McKinsey Journal. Today, a very interesting article from their newsletter arrived in my inbox.

It's probably hard to see how a hardcore business journal could have any applicability to backcountry skiing. Yet, after the recent events in British Columbia, I think an article on behavioural strategy is highly appropriate. I had to read the article several times to digest the information, but it's definitely worth the effort. The article touches on some topics that will be familiar to most people who have studied avalanche accidents: why do people make such poor decisions in the face of hard evidence? And why does it happen ... again and again and again?

The article also discusses the importance of the process used to make decisions, which for many recreationists is not formalised. Information from the article dovetails nicely with important behavioural research by Pascal Haegeli, Ian McCammon, and Laura Adams.

At the beginning of the 2009-2010 season, I wrote a short article on the strategies and tactics of safe winter backcountry travel. When traveling in the backcountry, we all have a set of tactics ( travel technique, snowpack analysis ) that we use to implement our ski trips. This begs the question: what's the strategy? And what procedures are used to uncover and neutralise biases in the strategies we use to reach our objectives? Consider the following model:

objective = strategy + tactics

Now consider that each term in the model is subject to biases that serve as possible sources of error. Have unconscious biases caused you to make an error in your choice of objective? Is Mt. Snoqualmie a suitable objective for current conditions? What about errors in your strategy, such as choosing to ascend Mt. Snoqualmie by Phantom Trees instead of Cave Ridge? Finally, what about errors in your choice of tactics? Are you openly discussing your tactical choices, such as route-finding options, with the other members of your party?

Since dangerous errors are possible at any level ( objective, strategy, or tactic ), it seems important to devise methods of uncovering errors for each level. Products such as the NivoTest, Werner-Munter 3x3 and the Avaluator are designed to help backcountry skiers choose appropriate objectives. However, these products do little to address the relationship between biases and the potential for error at the level of strategy or tactic. While it certainly is true that choosing an objective that is unsuitable for current conditions is likely to result in dangerous strategic and tactical errors, strategic and tactical errors can be just as dangerous even if the choice of objective is reasonable.

For Example: the backside of the Avaluator features obvious clues about snowpack instability, but I wonder if obvious clues about biases such as poor communication and failing to acknowledging uncertainty would be more appropriate? Neither the NivoTest, nor the Werner-Munter 3x3, address these, or other, dangerous biases in any manner whatsoever.

Further research is needed, but since the same biases can affect objectives, strategies, and tactics, I don't think it's terribly difficult to develop sound methods that address biases at all three levels.

Some interesting points from the article:
  • Daniel Kahneman has pointed out, the odds of defeating biases in a group setting rise when discussion of them is widespread. ( His work has been cited by many avalanche researchers. )
  • Particularly imperiled are senior executives, whose deep experience boosts the odds that they will rely on analogies, from their own experience, that may turn out to be misleading. ( This finding is perfectly aligned with research from Laura Adams. )
  • Pattern recognition is second nature to all of us—and often quite valuable—so fighting biases associated with it is challenging. The best we can do is to change the angle of vision by encouraging participants to see facts in a different light and to test alternative hypotheses to explain those facts. ( The Avalanche Handbook specifically discusses this as well. )
  • Most executives rightly feel a need to take action. However, the actions we take are often prompted by excessive optimism about the future and especially about our own ability to influence it. ( This has been discussed by avalanche researchers as well. )
  • To make matters worse, the culture of many organizations suppresses uncertainty and rewards behavior that ignores it. ( This has been discussed by avalanche researchers as well. )
  • Superior decision-making processes counteract action-oriented biases by promoting the recognition of uncertainty. ( Managing uncertainty is discussed throughout this blog as a fundamental element of safe winter backcountry travel. )
  • Stability biases also include loss aversion—the well-documented tendency to feel losses more acutely than equivalent gains—and the sunk-cost fallacy, which can lead companies to hold on to businesses they should divest. ( Or, loss aversion can cause us to continue toward a summit when we should turn around. )
  • An absence of dissent is a strong warning sign. ( Poor communication. )
  • To support those new norms, we also need a simple language for recognizing and discussing biases, one that is grounded in the reality of [mountain] life, as opposed to the sometimes-arcane language of academia. ( Avalanche education could use this as well. )
I realise that an article from The McKinsey Journal could seem highly out-of-place in a blog about avalanche safety. Yet, after a second and third read, it's easy to see how dangerous and costly biases become embedded in our decision-making procedures.

This is made all the more difficult by the fact that decision-making errors don't matter when the snowpack is mostly stable, since the chance of triggering an avalanche is relatively low. Over time, as the article points out, these biases become embedded in our procedures and errors arise ... then what happens?

Well, you get to decide.
My next post discusses the 7 Deadly Sins of Winter Backcountry Travel.

Tuesday, March 2, 2010

Does Size Matter?

In a big country dreams stay with you—Big Country

Like people, terrain comes in a bewildering variety of size and shapes and we all have different preferences. Sometimes terrain is obviously dangerous, and other times terrain is obviously safe. At the beginning of the season, I wrote a post about perception of instability. Like instability itself, perception of instability is rarely static. This means that your beliefs about instability change with place and time.

According to The Avalanche Handbook, variations in perception are usually very low when instability is widespread. On the other hand, variations in perception are most frequent, and most dangerous, when instability is not widespread, but not exactly scarce, and when the energy required to release an avalanche is relatively high.

The same rules apply to terrain. Perception of hazard is much better when terrain is obviously dangerous. Perception of hazard is less important when the terrain is actually very safe, since accidents are less likely. Serious errors are most likely when people choose "safe" terrain that sits in the middle of the spectrum. This is a fairly strong claim, but I think it's well-supported by records of skier-triggered avalanches accidents in North America.

Figure 1.1. Scale of terrain evaluation. Terrain is either very safe, very dangerous, or somewhere in between. Where does your secret stash fit on this scale?

Evaluating Terrain
Review the following images and think about where the images belong on the terrain scale. Where are you most likely to travel when snow quality and avalanche hazard are high? Where are you most likely to let your guard down?

Figure 1.2. Mt Currie, British Columbia. COMPLEX terrain featuring exposed alpine features and numerous avalanche paths. Many of these avalanche paths fall more than 1000 metres.

Figure 1.3. Twin Sisters, Washington State. COMPLEX terrain with numerous avalanche paths. Many of these avalanche paths fall more than 1000 metres.
Figure 1.4. Mt. Snoqualmie, Washington State. COMPLEX and CHALLENGING terrain with many safe options. This is exactly the type of middle ground terrain where backcountry skiers are at high risk for perceptual errors. There are many options to reduce exposure, and just as many options for small and medium sized avalanches during poor conditions. 
Figure 1.5. Central Cascades, Washington State. CHALLENGING and SIMPLE terrain. As above, there are many options to reduce exposure, and many options for small, but deadly, avalanches.

Why Size Matters
Of all the images, can you take a guess where the most accidents, burials, and fatalities have occurred? The answer is the last image, and it explains why size matters. ( Accessibility is part of the problem as well, but if i can believe what I read on the Internet, many local backcountry skiers believe that the terrain in the last image offers safe options during high avalanche danger. )

Size matters because small avalanches happen much more frequently than large avalanches, especially with respect to skier triggering. Large natural avalanches are relatively rare because enormous slabs can stabilise under their own weight. Large skier-triggered avalanches are rare because the application of dynamic forces from skier influence does not extend much deeper than 1 metre.

However, skiers like to choose terrain in the middle of the spectrum, especially on days when snow quality, and avalanche hazard, are high. Canadian researcher Laura Adams refers to this technique as satisficing: some desirable attributes, such as large, open slopes, are sacrificed while others, such as steepness, are preserved. This technique is definitely valuable, but it has dangerous side effects that contribute to avalanche accidents: choosing terrain in the middle of the spectrum makes it easy to rationalise and let your guard down. The line of thinking is as follows: we sacrificed the big open slopes up in the alpine in favour of steep slopes with trees. The only problem here is that the snowpack and the laws of physics simply don't care about your sacrifices.

The second problem is that most people are killed by avalanches that are, in the big picture, basically tiny. A complete burial requires very little snow, and most slopes that are fun to ski hold enough snow for several bad accidents. These "safe" terrain choices are even more dangerous because most middle ground terrain has at least some tree coverage. A recent study showed that skiers in North America are much more likely to die from traumatic injuries due to trees and rocks, than their counterparts in Europe.

Where I ski in the Washington Cascades and throughout British Columbia, most skiers regard a small slope as ~300 vertical metres, or about 1000 vertical feet. Many popular locations for day trips and yo-yo skiing involve such slopes. At the smallest scale, snowpack information becomes very important: terrain answers the question of is an avalanche possible, and snowpack answers the question of is an avalanche likely. Since it's just as easy to trigger an avalanche on a small, steep slope, it's easy to see why snowpack evaluation can become critically important when small, steep slopes are chosen.

Terrain analysis is one of the best tools for managing uncertainty, but only when slope angles are managed very strictly. Very often, I read trip reports where small, steep slopes are chosen as safe options during times when avalanche hazard is high. This is a fairly poor application of travel technique because small, steep slopes can be just as dangerous as large, steep slopes-especially if you don't seek out specific information about the snowpack.

What are your beliefs about the likelihood of avalanche formation on small, steep slopes? Do you believe small, steep slopes are safer? The record of avalanche accidents in North America provides a clear and resounding answer.

Accidents Involving Small Avalanches