Introduced in Europe during 1993/4 winter the five level avalanche danger scale was adopted, with minor changes, in America and Canada the following year. Despite criticism of the five levels and their interpretation the advantages of having a single scale across different countries are evident.
In 2005 a working group of leading North American avalanche forecasters began an update of the scale. The first aim was to produce a single definition of avalanche hazard, danger and risk for use in Canada and the United States. However it was also apparent that users of the bulletin had difficulties understanding the danger levels. The revision aims to improve the scale as a public communications tool for different classes of user from forecasters through to recreationists.
The avalanche danger scale accompanies avalanche bulletins and provides a relative measure of avalanche danger over a period of time for a given area. It is a matrix of danger levels and visual and textual interpretations of those levels.
During the design process the team carried out extensive end user testing including a two week period in March 2009 where Ipsos/Reid interviewed 4423 participants to evaluate scenarios in relation to the danger rating and their ability to choose appropriate courses of action. There was also extensive liaison with European avalanche forecasting services in order to harmonize the terms used. The new scale has been translated into French for Canadian use.
Since 2004 the European and North American avalanche warnings have evolved to a tiered approach in order to deliver information appropriate to different types of user, from simple danger levels and keywords to more complex travel advice. The new scale builds on this approach.
Numbers are used to indicate the danger level. These are important where the users are multilingual, such as in major European ski resorts or Canada. The revised North American scale includes numbers 1 (Low) to 5 (High) matched to each signal word. There are issues with their interpretation.
The avalanche danger doesn’t grown in a linear fashion, the numbers 1-5 would suggest that level 3 is an average level and conditions are “not that bad” whereas there is a significant jump from level 2 to level 3 etc. Research by Werner Munter suggests a doubling of risk between each danger level giving a statistical scale of: 1, 2, 4, 8, 16 (as used in Munter’s risk reduction method). Other researchers believe the increase in risk is even more marked. Although the progression is obvious, using a good signal word conveys more meaning about the conditions.
Low, Moderate, Considerable, High and Extreme… they are not sufficient on their own to interpret the risk but a number of decision making frameworks have evolved which depend on these levels. Low, Moderate, High and Extreme are obvious but the same cannot be said for Considerable which continues to demonstrate its ineffectiveness at communicating danger, around half of all French and Swiss fatal avalanche incidents occur at this level. This term is subject to wide variation in comprehension as the Ipsos survey demonstrated.
The Norwegians proposed Low, Moderate, High, Very High and Extreme. However this idea has not been adopted by the North Americans as it would dramatically redefine the terminology leading to further confusions by end users.
Colors are another method of communication and they are particularly effective on maps, signs and web sites. The colors used on the old scale remain with minor variations, for example for web use.
Parks Canada introduced a tiered warning system in 2005. The French have a similar system with their “Carte Vigilance’ with a Green, Yellow, Orange, Red scale. This scale is aimed at users with no specific knowledge and is communicated on national news bulletins. On the French scale orange roughly equates to a risk 3/4 and red a risk 4/5. The Swiss use 4 warning icons based on the same colour scheme as the French.
To standardize with the North American effort and the Parks Canada initiative the Swiss modified their icons and added a fifth, linking them directly to the five level avalanche scale. This system was adopted by the European Avalanche Warning service in 2009 as a common standard and will be used in the North American system.
This has been significantly improved to give strong, unambiguous statements on what the conditions are and how to travel safely. The statements, such as “Dangerous Avalanche Conditions” at level 3, are designed to be clear and concise. This column can also help avalanche forecasters determine the danger level by asking themselves the question “what travel advice would I give today?”
Likelihood of avalanches
This is the chance of an avalanche occuring, both natural and human triggered. The term “probable” has been eliminated and the wording corresponds much more closely to the statistical risk at each level (see Numbers, above). The following five terms are now used: Unlikely, Possible, Likely, Very Likely and Certain. For forecasters this is the most important column for determining the proper rating and there was always much debate over the words Possible and Probable. The new terms, combined with the avalanche size provide less wiggle room.
Avalanche Size and Distribution
Avalanche danger is defined as a combination of likelihood, size and distribution. This size of an avalanche has a direct bearing on the consequences of triggering an avalanche. To give an example, the likelihood of small avalanches on open terrain with gentle run-outs has less severe consequences compared to where there are terrain traps or obstacles such as trees or cliffs.
Giving definitions of avalanche size and distribution linked to the danger levels is a new and important piece of information for all audiences.
The new scale will be used from winter 2010/11 in Canada and the United States and from the winter 2011 in New Zealand. It provides a clearly defined avalanche hazard, danger and forecasting model and is based on best available information on snow science and risk communication.
This article is based on: North American Public Danger Scale Revision, Grant Statham; The Avalanche Review, Vol 29: no.1. October 2010.