Managing Avalanche Uncertainty

"It ain’t what you don’t know that gets you into trouble. It’s what you know for sure that just ain’t so."—Mark Twain

Backcountry skiers thrive on the unfamiliar: new places, new challenges, new skills and the feeling of uncertainty that goes with it. While uncertainty is integral to great adventure, it also creates the avalanche problem.

Uncertainty is a state of mind associated with a lack of knowledge about something. You are uncertain about everything in the future, but skiing in avalanche terrain takes uncertainty to a higher level.

"You can’t just live in a padded box. I don’t want to."—Wendy Wagner

Snowpack varies so much over mountains and over single slopes. At ISSW 2016, Canadian Mountain Holidays guide Todd Guyn said, "Due to the spatial variability and the physical environment of the mountains, we often work in a highly uncertain state." You dig a pit here and it’s different over there. Conditions change every day.

In addition to snowpack variability, avalanche danger is complicated by incomplete snowpack information. We do our best to collect pre-trip information and make field observations, but this information has gaps. We don’t know exactly what is happening within the snowpack.

How can we make good decisions that reduce avalanche risk in the face of so much uncertainty? Here are five key actions that can help us incorporate a more objective approach to managing uncertainty in our decision-making process.


The first step to managing uncertainty is to train yourself to think and speak in terms of uncertainty rather than certainty. Initially, this is a difficult concept to wrap your head around. The trick is to resist the human tendency to feel more certain about something than conditions warrant. Practice seeking information that counterbalances this tendency, just as we train ourselves to make instability tests for unstable snow. Get comfortable saying, for example, “I don’t know, what do you think?” and “This storm snow worries me.” Admitting that you feel uncertain is difficult for many people. It might even deflate your manliness a bit. In the end though, communicating uncertainty allows us to work better as a team and to make better decision about where to ski.


The purpose of making snow and weather observations in the field is to reduce uncertainty. These observations allow us to make better decisions. The best indicators of unstable snow are red flags. Observing any of these red flags is evidence of elevated avalanche danger. For example, if we see an avalanche, we know the snow is unstable. This reduces our uncertainty about what to do. We know it’s time to seek route option B or C. The opposite is also true. Lack of red flags doesn’t mean you are safe. In The Avalanche Podcast, episode 7, Don Sharaf says “When evidence is rare, uncertainty is high.” If you don’t see red flags, you need to observe snow conditions closer for evidence of avalanche problems.

This ‘16/’17 season in Southcentral Alaska, with its shallow snowpack, the ski pole probe test was an effective way to observe the persistent slab avalanche problem. Simply push your upside-down pole into the snow. Feel for a strong layer over a weak layer. Each test takes seconds, allowing many ski pole probes to be made throughout the day.


Recognize that a deep slab avalanche problem is a high-risk, high-uncertainty situation. Deep slabs form when a weak layer of snow is covered by a slab one meter or more thick. The depth of the weak layer makes it difficult to detect the presence of this problem with a ski pole probe or even a pit. The advisory is often the best indicator if a deep problem exists. A deep slab problem can show no signs of instability (evidence is rare, uncertainty is high). Even with numerous skier tracks, nothing may happen. But if one skier hits the wrong spot, such as a thin portion of the slab, the gorilla may wake up, and it will be pissed. It’s a low probability, high consequence situation. In the backcountry, you and I have almost zero experience with these gigantic deep slab avalanches.

In early February 2017, the Turnagain Pass avalanche advisory (below) showed moderate avalanche danger because of a deep slab avalanche problem. This danger rating is known as Scary Moderate. Uncertainty was high, so avoiding avalanche terrain in this zone was the solution. Yes, people got away with skiing big lines during this deep slab cycle, but it’s Russian roulette. You can accept the risk or you can chose to not poke the sleeping gorilla.

Below is the advisory a few days later. The advisory text describes the high danger was from a storm slab problem. Two glaring red flags were present: recent heavy precipitation and high wind. Because avalanche danger was high, and evidence was plentiful, uncertainty was low. We avoided the gorilla that was wide awake and beating its chest.

Wendy Wagner, Southcentral Alaska’s avalanche authority, summarizes reliability of information about the nine avalanche problems in her paper Avalanche Problem Toolbox. In her graph of Reliability of Tests and Observations (below), unreliable translates to uncertainty. Tests and observations on deep slab, wet slab and glide avalanche problems are unreliable, so uncertainty is high.

“Are you going to draw the line with a fine-tipped pen and stand with your toes dangling over the edge, or are you going to draw it in fat Sharpee and take three steps back?”—Karl Birkeland


Karl Birkeland, director of the National Avalanche Center, suggested drawing a wide line. On one side you know you will trigger an avalanche (you have low uncertainty). On the other side of the line you are pretty sure you won't trigger an avalanche (you also have low uncertainty). If you operate too close or within that line, your risk of getting caught and possibly killed by an avalanche goes up. Birkeland suggests staying at least three big steps back from that line on the low-risk side. You may still get surprised once in a while, and you won't eliminate all risk, but you will minimize it. When uncertain, make it a habit to move further away from that line than you might initially think.

For example, where do you stop after a run? You might think you are in a safe zone, beyond the runout of the current avalanche problem, but are you? Even terrain is not certain. Chugach Powder Guides head honcho Henry Munter, calls this terrain uncertainty. You don’t know how big a slope can avalanche, so your judgement and intuition don’t work. Try to move further away from the runout—three more steps—than you might initially think.


Last, and most important for decision making is to listen to your uncertainty. When you feel uncertain, do something about it. For example, move to lower angle terrain or turn around. Even if you don’t have a concrete reason for feeling that way, it’s worth listening to your intuition. The few accidents I’ve been involved with were all preceded with a bad gut feeling, a feeling of uncertainty that I didn’t act upon. Since then I’ve trained myself to recognize when I’m in doubt: “Oh, I’m wondering too much. I better back off.”

* * * *

Shifting your mindset to recognizing and reducing uncertainty will take much of the confusion out of the snowy mountains. It did for me. Now it feels okay that I don't know. I'm not supposed to know. Instead I try to quantify my uncertainty as high, medium, low and use that to help make better decisions about where to ski.

Joe Stock

Author: Joe Stock

Joe is an IFMGA Mountain Guide with a passion for mountain adventure in Alaska. He has been climbing and skiing around the world for over 30 years with significant time in New Zealand, Australia, Asia, Alps, Andes and western North America. The mountains of Southcentral Alaska are his favorite. Joe has an undergraduate degree in geology and geography from the University of Canterbury in New Zealand and a graduate degree in watershed science from Colorado State University. The second edition of his guidebook for backcountry skiing in Southcentral Alaska, The Alaska Factor, was published in 2016. Joe lives in Anchorage with his wife Cathy and their Really Bad Orange Cat.