Skip to content

Cookies 🍪

This site uses cookies that need consent.

Learn more

Zur Powderguide-Startseite Zur Powderguide-Startseite
safety topics

Observation and assessment - what is important?

PowderGuide visits the SLF

by Lea Hartl 12/17/2017
Around 60 people are sitting in the seminar room of a Davos hotel. They include mountain guides, safety officers from municipalities and cantons, natural hazard experts from the Swiss Federal Railways and blasting experts and slope maintenance staff from various ski resorts. Lukas Duerr, avalanche warning officer at the SLF, asks the group: "What does serious danger actually mean?"

Murmuring. Minor additional load, widespread danger spots...

Duerr asks everyone to stand up and come forward. He hands out index cards with key points.

There are 25 cards, 5 for each of the 5 hazard levels. On the cards are sentences and keywords as used in the bulletin to describe the hazard levels. Some cards contain key data on the frequency of use of the different levels, or SLF recommendations for behavior, as they also appear in the bulletin. These are the linguistic expressions that the SLF uses to communicate the hazard levels on the one hand, and on the other hand, at least in part, to determine which hazard level prevails at all.

Duerr asks everyone who thinks that their card fits level 1 to stand in the right-hand corner of the room. Level 2 should stand next to it, level 3 again next to it and so on.

For all those who also want to play along, we have recreated this virtually here. Results appear when you click on Submit at the end and then on "View results".

safety topics
presented by

The common language as a basic requirement for avalanche warning

Unlike in Austria, for example, the avalanche situation report for the whole of Switzerland is compiled centrally in Davos. The avalanche forecasters in Davos can get an approximate overview of the situation based on weather station data, but the assessments and feedback from the human observer network that the SLF maintains throughout Switzerland are at least as important as the automatic measurements.

There are different types of SLF observers who generate different data: the flat field observers measure fresh snow and snow depth on a daily basis, make various observations to assess the avalanche risk and regularly dig profiles in staked-out measuring fields in the plain. The main aim here is to continue valuable, long measurement time series. In addition, in extreme situations, when it is not possible to carry out snowpack surveys elsewhere, the flat field profiles can still be used.

Selected mountain guides report their observations from the field via a specially developed app, which they can use to communicate their assessments to the SLF and exchange information with each other. In addition to general observations, for example on danger signs or weather developments, the observers record layer profiles, carry out stability tests and submit a risk assessment. Ski resort employees and avalanche services also report observations and hazard assessments from their regions.

When it comes to hazard assessments in particular, it is not so easy to express exactly what you want to say on the one hand and to interpret other people's statements correctly on the other. What level would I give today? Why? And, does my assessment agree with the bulletin? If observer A reports, "today it is dangerous", he may mean something completely different from observer B.

In a study by SLF researchers Frank Techel and Jürg Schweizer, these differences were analyzed in more detail: the evaluation of 10,000 local hazard level assessments by SLF observers shows that the observers disagree with each other in 22% of cases, even at close spatial distance. In 76% of cases, the local hazard level assessments of the observers agree with those of the bulletin, although there are clear differences between individual observers - some consistently rate the hazard higher or lower than others. There are also certain differences between different groups of observers. For example, flat field observers and ski resort employees tend to rate the danger higher than observers who are themselves out and about in the open ski area (mountain guides, SLF employees).

In order to counteract such differences, the participants of the SLF training course "Observing and assessing", all of them observers for the SLF, now stand in the seminar room with their index cards.

Some go purposefully to their seats, others hesitate, consult, compare the cards and then join one of the groups. In the end, there are significantly more people on the left-hand side of the room, at the lower danger levels. Only a few feel they belong to levels 4 and 5.

Course leader Lukas Duerr goes through the cards one by one until the groups have arranged themselves correctly and there are 5 people at each hazard level. The general conclusion from the exercise is: "Not all that clear, but logical at second glance."

The exercise makes it clear how important a uniform understanding of the hazard level definitions is and underlines Duerr's request to always justify hazard assessments in as much detail as possible and to find a uniform language for this. Today it is dangerous" becomes "today it is more dangerous than yesterday because fresh drift snow has formed. The following alarm signs could be seen..." Those who issue hazard assessments, including the hazard level, might think more carefully in future about the difference between "typical" and "frequent", or between "avalanches are possible" and "avalanches are to be expected".

Skier foursome vs. traffic foursome

The next item on the agenda shows that the understanding of danger levels can also change over time. Duerr presents a map of the Alpine region, showing how often the major danger level is used in the various countries and warning regions. Switzerland appears to be a kind of island. Major is assigned relatively often in France in particular, but most other Alpine countries also use level 4 noticeably more frequently than Switzerland.

These differences cannot be explained by differences in the snow cover alone, but are due to the different perceptions of the meaning of the level. The use of danger level 4 will certainly also occupy the EAWS (European Avalanche Warning Services) group in the near future.

It remains to be seen whether and how quickly it will be possible to use a similar hazard level 4 throughout Europe. In any case, Switzerland and the SLF have more or less explicitly distinguished between the pure "traffic level", which by definition must include the endangerment of infrastructure, and the "skier level" since this season. The latter is to be awarded in future if roads and exposed buildings are not necessarily at risk, but skiers are particularly affected. This could be the case, for example, if the trigger readiness is very high and the danger spots are very numerous, but the avalanches are not large enough to reach roads or buildings. Previously, in such cases, a 3+, the dark orange three, was awarded in Switzerland.

In the language of the index cards, the change concerns situations in which "spontaneous avalanches of medium size are typical", or in which "remote triggering is typical and people can trigger small and medium avalanches very easily, even if the spontaneous avalanche activity is small".

From the classroom to the snow

The next morning, we head outside in small groups. It has snowed overnight: what makes for a picture-postcard winter idyll in the village has been blown away on the Weißfluhjoch. There is hardly any snow left on wind-exposed terrain peaks, but all the more in gullies and hollows in the lee. The danger level for the Davos area today is moderate, further south it has snowed more and the bulletin gives it a three.

During a short ascent, Duerr encourages us to look at the area and discuss what we see. He explains the surrounding peaks and popular Davos freeride runs to the fresh SLF interns and me, but he actually wants to hear what we think of the snow. The avalanche warden's gaze doesn't (or, in the case of avalanche warden/mountain guide/skier Duerr, not only) linger on the beautiful lines in the surrounding area, but on snow-covered and fresh crack edges, wind-joints and dunes, rough ice on piste boundary posts, sliding snow cracks in the steep south-facing slopes and the melt crust that the wind has exposed again.

Whether we agree with the two-man today? Hmmmm. It was already very windy, and the fresh snow... On the other hand, there are no clear signs of alarm, so it will be okay.

safety topics
presented by

We start looking for a suitable profile location. That's not so easy either and we discuss for a while whether we dare to venture into a relatively large, blown-in slope. At some point, everyone realizes that if you have to discuss it for so long, you'd better not dare. Yesterday in the "classroom" the observers were told: Please don't get buried at all, and certainly not when digging profiles for the avalanche warning service. There have already been such cases and we still have the images of accidental avalanches vividly in our minds. We descend one by one and find a better spot further down.

After recording the layer profile, an ECT is dug. The whole block breaks on the 15th shot. The subsequent rustic block comes off as a whole when SLF trainee Dylan carefully climbs on top of it on skis. We're glad that we decided against the bigger slope further up, but we're not so sure about today's two-pointer.

In the afternoon, we compare our observations with those of the other groups. Apart from us, only one group was able to release their slide block completely. All the others had partial breaks or no result at all. This comes as a surprise to us. Our slide block was rather scary and afterwards we all had the feeling that we should be a bit more careful than we might have been if we had only registered the two in the bulletin.

If you force yourself not to attach more importance to your own, not so great, sliding block than is appropriate when assessing an entire region and look at all the observations as a whole, it quickly becomes clear: yes, the deuce was okay. Our slide block result and those of the others match the index cards "Alarm signs can occur sporadically", and "Avalanches can be triggered especially on very steep slopes ....", or the definition of level 2 according to the danger level scale: "The snowpack is only moderately consolidated on some steep slopes, otherwise generally well consolidated".

For me personally, our landslide block result and the subsequent discussions in the groups and with the course leaders led to the following realization: "can occur sporadically" and "only moderately consolidated on some steep slopes" are real descriptions of the situation, which can have real consequences. If we follow Munters reduction method, this is the residual risk, which can be minimized but not completely eliminated. When we encounter the residual risk in the form of an unfavorable slide block or even an avalanche, we (at least I do) always find it somehow surprising - "such a slide block on a two-man, that can't be!" Yet the whole system is built precisely on the probabilities that formulations such as "isolated" and "some steep slopes" describe.

It quickly becomes clear that neither a single observation nor the danger level alone are sufficient for an individual slope assessment, especially for the Moderate and Significant levels (with Large and Very Large, you usually don't even get into embarrassment). Fortunately, the avalanche bulletin contains more than just the danger level! Firstly, to perceive this "more" information in the morning powder fever, to verify it in the terrain with all possible observations and then to behave accordingly is a challenge that no avalanche warning service in the world can take away from us as skiers.

PowderGuide would like to thank the SLF and especially Lukas Duerr for the opportunity to take part in the training course "Observation and Assessment"! We hope we can come back again sometime!

Photo gallery

ℹ️PowderGuide.com is nonprofit-making, so we are glad about any support. If you like to improve our DeepL translation backend, feel free to write an email to the editors with your suggestions for better understandings. Thanks a lot in advance!

Show original (German) Show original (French)

Related articles

Comments

safety topics
presented by