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

Friday, January 22, 2010

Backcountry Recipe Book

Terrain + Snowpack + Weather + Human = Uncertainty

NOTE: I was originally going to write a post about map reading, but I've been too busy with the non-digital part of my life, which includes boring tasks such as family chores, earning a living, and somehow finding time to play in the snow. In any event, I haven't been able to wrap my head around the topic of map reading just yet. This post discusses decisions and does not cover other important elements of the sport, such as trip planning and travel technique. { This post was revised on November 18th, 2010. }

I've written some complicated posts about backcountry avalanche forecasting, snowpack instability, terrain, snowpack, and weather. Regardless of complexity, the essential question remains: how do you leverage all these observation techniques to produce a backcountry avalanche forecast that enables you to make sound decisions?

Learning which observations have priority and correctly interpreting observations is required for successful backcountry avalanche forecasting. Priority of observations ( relevance ) is usually determined by the size of the forecast area, but interpretation of relevant observations ( the relationship, or link, between the observation and snowpack instability ) is much more subjective and sometimes quite indirect.

Avalanche experts recommend integrating observations whenever possible instead of relying on disconnected observations. Yet the process of integrating observations is complex and somewhat mysterious. In this discussion, we're going to examine the linkages between types of observations, and develop a list of simple keywords that help frame these complicated problems. NOTE: The process of integrating observations does not apply to the golden rule: alternate plans are required if you observe absolute signs of instability such as whumpfing, cracking in the snow surface, or recent avalanches.

Let's discuss a recipe for decisions, and along the way, I'll restructure the typical decision-making process to explicitly revolve around the balance between awareness and uncertainty.

A Recipe For Decisions
I wrote an article about avalanche education for the December 2009 issue of The Avalanche Review, and in the article I discuss the importance of teaching students how to make decisions that blend uncertainty, safety, and fun. This structure for decisions is compatible with realistic human behaviour, the importance of which is discussed in The Avalanche Handbook.

Recently, there have been a number of online discussions about decision-making, some of which have included lively conversations about backcountry outings during which avalanches were triggered by members of the party. From these discussions it has become increasingly clear to me that the lack of a general, universal recipe for safe decisions is a primary source of confusion for recreational backcountry skiers. Rather than focusing on checklists or graphs, I am going to write about a simple approach to decision-making that integrates critical human elements into the mix.

From my post on TetonGravityResearch's Slide Zone Forum:

"Your question(s) underline a very common, and often unacknowleged, source of concern for anyone in the backcountry: Even after observing the terrain, snowpack, and weather, how do I manage the remaining uncertainty?"

Recreational backcountry skiers are taught, whether by other skiers or in a classroom setting, to use multiple observations before deciding whether or not to travel on a particular slope. Observations rarely provide clear and precise answers because the winter snowpack is conditionally unstable most of the time.
The standard recipe for decisions is as follows:

Terrain + Snowpack + Weather + Human = Decision

From my own experiences, including observations of recreational backcountry skiers, both online and in the field, I think it's fair to say that many skiers have difficulty using this recipe to connect their observations with their decisions. The difficulty in connecting observations to decisions revolves around what recreational backcountry skiers expect to gain from their observations. If I can believe what I've seen in avalanche safety training courses, and what I've seen in the field, the expectations of many recreational backcountry skiers are as follows:

Terrain + Snowpack + Weather + Human = Answers

Anyone with enough knowledge and experience knows that the observations rarely provide answers because the prevailing state of the winter snowpack is conditional instability. This basically means mostly stable with a chance of avalanches. Therefore, the recipe that ends with answers doesn't really exist, unless you are willing to accept answers that are very vague and seemingly unrelated to the situation at-hand. Most people won't find these "answers" very useful if they are expecting something definitive.

If we dig deep into both the scientific and human aspects of the problem, including a lengthy foray into stacks of published research, a far more intuitive recipe for decisions emerges:

Terrain + Snowpack + Weather + Human = Uncertainty

The key here is learning how to think about observations: observations are undertaken for the sole purpose of minimising uncertainty. Observations are not made for the purpose of finding definitive answers. The spatial and temporal variability of the winter snowpack ensure that some degree of uncertainty is always present. Therefore, in many situations, the best we can do is acquire information that minimises uncertainty before making the decision to travel on a particular slope. Observations rarely make the uncertainty go away entirely, and when they do, it is only when signs of absolute instability, such as cracking and whumpfing, are observed. In those situations, you can say the snow is absolutely unstable, but everything else is gray.

In many avalanche safety training courses, the element of uncertainty is often wrapped up in vague or complicated language about risk management or decision-making, yet such language is more suitable for insurance companies than backcountry skiers. The revised recipe presents a human-compatible method for making decisions based on uncertainty, and this method actually makes sense to most people.

Here's why:

As I said in my article for The Avalanche Review, during the process of purchasing an automobile, you don't acquire information to make a decision, you acquire information to reduce your uncertainty before making a decision. At least, this is how it works for most people. Rather than fighting against the inherently confusing uncertainty of observations, recreational backcountry skiers should be taught to use this uncertainty to their advantage.

Working With Uncertainty
It's safe to say that the amount of uncertainty present in any situation is highly related to the knowledge, skill, and experience of the observer, and it's also safe to say that these variables are influenced by the complexity of current conditions. According to research published by Laura Adams, the correct method for addressing this uncertainty is as follows:

Acknowledge + Identify + Reduce = Managing Uncertainty

The first step is to consciously acknowledge uncertainty. I find it helps to say it aloud: "What is my uncertainty?" The next step involves identifying the sources of uncertainty and deciding whether or not reviewing existing information or seeking additional information will reduce uncertainty.

Many times, it is not possible to reduce uncertainty by seeking out additional information, and this is exactly when many serious mistakes are made in the backcountry. A highly uncertain mind is very susceptible to unscientific speculation, dangerous biases, made up facts, and other forms of outright stupidity such as "explaining away the uncertainty" or "ignoring the uncertainty" altogether. With this in mind, it certainly seems reasonable to wonder if many grave but common errors, such as trying to "outsmart" the snowpack, are simply poorly managed reactions to high uncertainty.

Even with meticulous efforts to maintain awareness through careful and objective observations of the terrain, snowpack, weather, and human factors, recreational skiers should take an immediate step back when high uncertainty is present during decision-making. In general, without expert decision making skills, decisions made by recreational backcountry skiers should retain a conservative character when uncertainty is high.

The character of the correct decision can be easily uncovered by using observations to determine the level of uncertainty prior to decision-making. Failing to proactively manage uncertainty, including failing to recognise that uncertainty is always present, is a primary source of avalanche involvement and accidents in recreational backcountry ski outings.

While important, awareness and uncertainty are just two parts of the multi-faceted discipline of backcountry avalanche forecasting. Backcountry avalanche forecasting has other elements and relying solely on awareness and uncertainty will eventually lead to mistakes.

A minimalist framework for backcountry avalanche forecasting is discussed here.

Friday, January 8, 2010

Weather Observations: Trend Analysis

( 1 + x ) × ( 1 + y )—Ms. Steffens

NOTE: Backcountry avalanche forecasting is composed of four interconnected elements: goal, people, awareness, and uncertainty. Four types of observations, human, terrain, snowpack, and weather are used in backcountry avalanche forecasting. { Edited on March 7th, 2010. }

Backcountry avalanche forecasting relies on observations of the terrain, snowpack, and weather inside a single valley or backcountry ski run. Observations of terrain, snowpack, and weather are made to increase awareness and manage uncertainty.

Weather Observations: The Direction Of Instability
The primary difficulty with weather observations is the quantity of data, wide area of influence, and the uncertain relationship between observations and instability. Despite these complexities, weather observations focus on discerning the presence of, and effects of, three key factors that create and influence snowpack instability. These factors are listed below in order of importance relative to snowpack instability:
  • Loading
  • Warming
  • Cooling
Warming, cooling, and loading are trends that can increase or decrease snowpack instability depending on the current structure of the snowpack and the magnitude, rate, and duration of the trend across space and time. Individual variables such as cloud cover, slope aspect, local terrain, weather patterns, and radiation balance contribute to these factors, but ultimately warming, cooling, and loading are the practical concerns.

Warming, cooling, and loading create new snowpack instabilities while simultaneously altering the current snowpack in ways that can increase or decrease instability. A primary source of confusion is the circular nature of trends related to these factors, especially the relationship between these trends and snow metamorphism. Temperature changes are a good example. A warming trend can create instability by melting the snow and decreasing its hardness. However, when subject to a gentler warming trend, the direction of instability will eventually reverse itself because warm temperatures promote settlement and strength gains throughout the snowpack in general. Trends for cooling and loading also create circular variations in the direction of instability.

The direction of instability depends on the magnitude, rate, and duration of the trend relative to the current state of the snowpack. In simple terms, it is important to understand:
  • Magnitude. How much cooling, warming, or loading is expected?
  • Rate. How fast will the cooling, warming, or loading occur?
  • Duration. How long will the cooling, warming, or loading last?
You are probably aware that avalanche formation is much more likely at some times than others. Trends and their magnitude, rate, and duration are significant sources of uncertainty with respect to the formation of avalanches across space and time. These trends are responsible for the development of both long-term and short-term instability.

Table 1.1. Partial list of key weather observations mapped relative to instability and stability. Do you notice the potential for each factor to increase and decrease instability depending on the magnitude, rate, and duration? The circular nature of these trends, and their effects on both new snow and the existing snowpack, along with times when more than one trend is active ( cooling + loading ), are strong sources of confusion and uncertainty in backcountry avalanche forecasting. Make a list of your own and experiment.
LoadingMediumFast2 Hours+
LoadingLargeSlow22 Hours-
WarmingLargeFast18 Hours+
WarmingSmallSlow14 Days-
CoolingLargeSlow12 Days+
CoolingMediumFast4 Hours-

Table 1.2. Contributors to warming, cooling, and loading.
PrecipitationAir TemperatureAir Temperature
SkiersCloudsClear Skies

The current state of the snowpack, and magnitude, rate, and duration of current trends, are extremely relevant to backcountry avalanche forecasting. Are you preparing to descend a steep slope covered in fresh snow even though temperatures have been rising quickly for the past two hours?