Page 1 of 3 1 2 3 LastLast
Results 1 to 25 of 57
  1. #1
    Join Date
    Jan 2006
    Location
    Vanity Fair
    Posts
    2,621

    Do you use statistical methods of risk reduction?

    By statistical methods I mean Munter's reduction or 3x3 method and all the variations of this approach. I would be curious to hear what non-europeans think of it.

    In german speaking Europe, Munter (or variations thereof) has been taught pretty much as the be-all-end-all solution for avalanche risk mitigation for recreational users for 20 years. At least partially due to this, the avalanche danger level is all that registers from the bulletin for a lot of people (you don't need anything else to do the basic Munter calculation, although some additional information can be incorporated). In the most basic type of recreational avalanche course the focus would typically be on Munter, without mentioning anything that involves digging at all, or even discouraging people from digging pits. The argument is that most incidents can be avoided if you stick to Munter and everything else is so confusing for people that it is not beneficial. (which is true for a lot of people and a lot of settings)

    Recently there has been an increasing tendency to also look at other stuff again (i.e. dig and differentiate between types of danger, rather than just level of danger), at least at the higher end of the avalanche education spectrum (professional forecasters were always digging, of course). However, Munter is still mostly the standard that everyone is held to when determining whether something was overly reckless, in general and typically also in legal settings.

    In Austria we just had 2 high profile incidents with 6 fatalities, both guided groups, and both well within what Munter "allows". This has sparked quite some discussion about whether there is some kind of systemic flaw in how danger levels are assigned in the bulletin or, alternatively, a problem with Munter. A lot of people seem to think if something slides even though it was "allowed" according to Munter, the danger level must have been wrong. (For both incidents the danger level was clearly accurate in terms of the definition of the danger scale and the other content of bulletin is also above reproach, stating exactly where the problem areas are and what the problem is.)

    I am under the impression that Munter never really took off in North America, at least not to the extent that it did here. Is this true? Is it used at all? Is it taught and if so to what kind of user? What tactics for minimizing risk are taught at the most basic level of avalanche education?
    Ich bitte dich nur, weck mich nicht.

  2. #2
    Join Date
    Sep 2001
    Location
    Before
    Posts
    16,884
    Having done 2 level 1 and 1 level 2 where the level 2 was done at Silverton, a well regarded school in the states, I never heard or read of 'Munter'.

    Admittedly all those classes were quite a while ago, so things may have changed stateside.

    Pit analysis were touted as the best analysis in my classes. I had my doubts seeing as how snow quality can change over the distance of a few feet and I think that's been borne out in some field analyses of avalanche accidents.

    Another factor brought up in all of my training was exposure: where one skis, being aware of potential start zones and in particular terrain traps. I feel that's greatly overlooked.

    Personally, I focus on exposure and just a general vibe about how cohesive the snow is on the surface and how I feel. It's not really quantifiable.
    Merde De Glace On the Freak When Ski
    >>> Black Bamboo Sidewalled DPS Lotus 120 : Best Skis Ever <<<

  3. #3
    Join Date
    Jan 2009
    Location
    Squaw valley
    Posts
    1,528
    A statistical approach says, in average, you should expect a certain probability of an avalanche.

    Like the toss of a coin.

    The problem is that you don't have the luxury of averages.

    If ah avalanche happens while the chance was 10%, you're still dead.

    In my opinion, look, every day, at wind, temperature, precipitation, and form an opinion on whether there is a buried weak layer, the hardest factor to predict.

    Weak layer, don't ski, or ski low angle, and not below steep faces.

    And always dig a pit to verify.

    Then there are wind slabs, much easier to predict and avoid.

    It's more complicated of course, but imo, the buried layers are the ones where munter is dangerous.

  4. #4
    Join Date
    Jan 2006
    Location
    Vanity Fair
    Posts
    2,621
    Quote Originally Posted by rod9301 View Post
    A statistical approach says, in average, you should expect a certain probability of an avalanche.
    Like the toss of a coin.
    Your chances are a lot better than when tossing a coin if you do it by the book, but yes, it comes down to probability rather than a personal analysis of the snow pack. Munter and the others (Snow Card, Stop or Go,...) would not work without very good avalanche forecasts, which are of course based on snow pack analysis. It does simplify things a lot for the end user, but there are obvious limitations. And yes, I agree that those limitations are most apparent in persistent weak layer/ low probability-high consequence situations.

    Buster: thanks, that's what I thought. I find the difference in approaches interesting and am wondering why things turned out like this. Here is a kind of general, not very good explanation of what the Munter method is and here is an explanation of how to do the calculation. Here is a practical example. It is surprisingly hard to find english explanations online. Not hard at all in german.
    Ich bitte dich nur, weck mich nicht.

  5. #5
    Join Date
    Apr 2004
    Location
    Grenoble
    Posts
    343
    Quote Originally Posted by klar View Post
    By statistical methods I mean Munter's reduction or 3x3 method
    I was wondering whether 3x3 is really a statistical method or a decision aid? As the PisteHors link shows applying the reduction method, or any other statistical method, is a bit fraught with difficulty as the comments Rolf makes at the end of the article shows. It is probably best to err on the side of caution. What these methods do is keep you off > 30 degree slopes where there is a considerable or more risk on those aspects. Which I guess we could do by simply reading the avalanche bulletin.

    The 3x3 method is taught to French club leaders. I would only formally apply it if leading others as I'd want some way to justify my route choice if there was an accident.

    I think they are quite a good learning aid.

    I'd be interested if you can post the bulletins for the two Austrian incidents and maybe we can do the Munter calculations, I'm sure you are right but it could be a useful exercise.

    French guide Olivier Mansiot recently analyzed 40 incidents in detail with 4 factors taken into account as warning signals from the bulletin. He found that in 36/40 incidents there were 3 clear indicators that the slope being skied was dangerous. He also makes that point that risk 2,3 or 4 may be the same thing for at risk slopes. As guide Claude Rey once said "globally it is safe, locally there is a risk of death".

  6. #6
    Join Date
    May 2007
    Location
    Juxtaposition
    Posts
    5,818
    In Canada (and specifically tailored for Canadian conditions and resources): the Slope Valuation Card from the Avaluator. It highlights the interplay between weather, snowpack and terrain factors which are associated with increased probability and consequences. An excellent tool to at least remind users of what matters. Designed after an analysis of avalanche incidents over a long period of time in Canada.

    And no, it doesnt list slope aspect (north) as a factor.

    Via google... http://zacstracs.com/CAC-Outreach/331/slopeeval

    You can buy them from MEC.
    Life is not lift served.

  7. #7
    Join Date
    Jan 2006
    Location
    Vanity Fair
    Posts
    2,621
    davidof: I would say it is a decision aid based on an understanding of the probability you have of triggering an avalanche in a random slope for a given danger level. What Munter says is that if you stick with his calculation and do not go above a remaining risk of 1, you eliminate 99% of the risk, compared to someone who does not think about avalanche risk at all and does nothing to avoid it. As in, he defines a 1 in 100 chance of something happening as the cut off for an "acceptable" level of risk. You can stay below one if you want to be more conservative, or go above if you accept more risk.

    The bulletins are here: March 15, March 17

    for March 15 I would say it is: 4 / 2 x 2 (maybe 4) = 1
    danger level 2 --> potential 4, divided by steepest slope angle 35-39 (--> 2) and large group with distances (--> 2). For their route (not entirely clear where everyone was) you could probably argue that the angle was 30-34 and use reduction factor 4, which would put you quite a bit lower than 1. What it comes down to here is that according to Munter you need to look only at the slope angle of a fairly narrow corridor where you are skiing for moderate, but the whole slope for considerable. They apparently triggered it in a spot that was well below 35 and it propagated to steeper sections (beyond what you supposedly need to take into account for moderate) and then ripped out in the largest possible way.

    for March 17: 2 / 3 = 0.6 Low danger, small group skiing one by one. Actually I'm not sure now if you need a minimum of 2 reduction factors? I'd have to check. You could probably make a sketchy argument for using reduction factor "not on the slopes the bulletin specifically mentions", i.e. 4 if a second one is required.

    Especially for the 15th (huge slide, 4 dead) the bulletin does clearly state that the kind of slope they were on is dangerous (northern sector, not skied much, above 2200m). If you do only the calculation you can tune it to make it work, though not in the cleanest possible way. You could definitely argue that they were pushing it with the tuning but that's pretty much how people use it, because the system allows you to. For the 17th the bulletin is a little more ambiguous for the place where the slide happened. It was a much smaller slide on a small, isolated terrain feature and they were caught in a terrain trap.

    March 15th:


    March 17th:



    neck beard: interesting, thanks!
    Ich bitte dich nur, weck mich nicht.

  8. #8
    Join Date
    Sep 2010
    Location
    Shuswap Highlands
    Posts
    1,535

    Do you use statistical methods of risk reduction?

    In a similar vein as the Avaluator mention by Neckbeard, BC Search & Rescue Association has developed a protocol called RADeMS to assist with risk mitigation for tasks:
    http://www.bcsara.com/projects/radems/

  9. #9
    Join Date
    Apr 2004
    Location
    Grenoble
    Posts
    343
    Quote Originally Posted by klar View Post
    davidof: I would say it is a decision aid based on an understanding of the probability you have of triggering an avalanche in a random slope for a given danger level.
    I was thinking about 3x3 rather than the reduction method. In 3x3 you consider global, local and zonal factors - planning, when you arrive and once on the tour. It is a lot more subjective than pure statistical methods. Global would be taking the Bulletin and Weather forecast, looking at the slopes at risk, picking a route choice that avoids these. Local - you arrive and it has snowed a lot more than anticipated, that fat bloke has turned up and he's always having accidents, does your route choice still apply. Zonal, you arrive at some key slope, do your route choices still apply, maybe there are more people on the route than anticipated, perhaps there is recent activity on similar slopes.

    Quote Originally Posted by klar View Post
    What Munter says is that if you stick with his calculation and do not go above a remaining risk of 1, you eliminate 99% of the risk, compared to someone who does not think about avalanche risk at all and does nothing to avoid it. As in, he defines a 1 in 100 chance of something happening as the cut off for an "acceptable" level of risk.
    A 1 in 100 chance wouldn't interest me too much, I'd be getting caught in avalanches practically every winter, as would you I expect. I thought the reduction method halved the likelihood of getting caught. That is, the number of avalanche deaths per average season would half.

    I'll have to go back and reread his book.

    In Europe there is the Nivotest and also Alain Duclos has proposed his method called "vigilance encadre" - which seems like the Zonal part of 3x3

    https://prezi.com/mf2uwyozajj0/avala...-et-vigilance/

    Thanks for the links to the Bulletins, I'll take a more detailed look.

  10. #10
    Join Date
    May 2007
    Location
    Juxtaposition
    Posts
    5,818
    Quote Originally Posted by klar View Post
    ...the 15th (huge slide, 4 dead)

    March 15th:
    "That aint no windslab."

    Quote Originally Posted by klar View Post
    In Austria we just had 2 high profile incidents with 6 fatalities, both guided groups, and both well within what Munter "allows". This has sparked quite some discussion about whether there is some kind of systemic flaw in how danger levels are assigned in the bulletin or, alternatively, a problem with Munter.
    Emphasis mine.

    - Does the bulletin specifically identify persistent or deep persistent slab problems when appropriate? And was it appropriate to do so on that day?

    - Consequently, does the Munter 3x3 calculation factor in the existence of said persistent problems?

    In short and at the foundational level, the existence of such problems = automatic no questions asked dial back the risk taking by a factor of X, regardless of aspect or elevation, for all recreational groups who use systematic methods of hazard assessments (like Munter and numerous others).
    Last edited by neck beard; 03-18-2017 at 06:34 PM.
    Life is not lift served.

  11. #11
    Join Date
    Apr 2004
    Location
    Grenoble
    Posts
    343
    Ok I'm just taking a look at the 15th March incident

    http://lawinenwarndienst.blogspot.fr...rubenkopf.html

    Correct anything I've got wrong, I have no clue about the area:-

    So the summit is at 2450 and the avalanche is shown on their map as running from 2400 to 2000 meters. That seems to tally with the photograph

    Name:  image002-759625.jpg
Views: 433
Size:  81.6 KB

    avalanche triggered on some old depth hoar probably formed in the thin snowpack at the start of the season, obviously not well bridged.

    I would use the following:
    Risk 2 Moderate on the slope they were skiing
    Slope > 35 degrees < 40 degrees
    North Sector Slopes Specifically Mentioned in the Bulletin
    Large Group

    So Risk is 4
    RF for slope is 2
    we don't have a second class reduction factor: it is a N. sector slope and is specifically mentioned in the bulletin
    I'd also say: large group with limited spacing as 4 members killed

    So that gives risk of 2 (it would have been 1 with better group spacing).

    Note: I've now gone back to your calculation. In my description of the Reduction method only the steepest section of the slope is mentioned, no reduction factors for moderate. I'm wondering if this makes sense. The problem we've seen in Austria and in France the last couple of seasons are buried weak layer formed at the start of the season due to thin snow-packs that extend over the whole slope. If you trigger in one spot, even on the flat, the weak layer collapse can propagate widely and if the slope is steep enough then trigger a big slide.

    http://www.avalanches.org/eaws/en/ma...er=basics&id=2

    I'm not sure that a risk of 2 / Moderate precludes this large kind of avalanche, just that they are confined to rare, steep slopes.

    I think this shows the danger: if you are looking for a reason to ski this slope, the Reduction method gives it, if you are looking for a reason not to ski it, well it does that too depending on how you apply the method.

  12. #12
    Join Date
    Jan 2006
    Location
    Vanity Fair
    Posts
    2,621
    Quote Originally Posted by neck beard View Post
    "That aint no windslab."
    Emphasis mine.

    - Does the bulletin specifically identify persistent or deep persistent slab problems when appropriate? And was it appropriate to do so on that day?

    - Consequently, does the Munter 3x3 calculation factor in the existence of said persistent problems?
    Yes the bulletin did specifically mention exactly what and where the problem was. Neither the 3x3 nor the reduction method say anything explicitly about this type of problem. Of course they encourage you to stay away from any areas that the bulletin calls particularly dangerous but they don't "forbid" it on principle. There is no explicit differentiation between types of danger (wind slab, persistent weak layer, etc), with the exception of a few special rules for wet snow.

    The second, smaller slide also was not a wind slab. First day with an isothermal pack, the layer it slid on was the same kind of old facets as with the big one.

    Quote Originally Posted by davidof View Post
    I think this shows the danger: if you are looking for a reason to ski this slope, the Reduction method gives it, if you are looking for a reason not to ski it, well it does that too depending on how you apply the method.
    I agree. Most people use the reduction method to find a reason to ski something, not to find a reason not to. I would also have to reread the book to be sure whether this is actually in the book or was added later by the Alpine Clubs for Stop or Go and the Snowcard, but the way it is taught is that you only need to apply whatever method you use to exactly where you are skiing for low danger, to a 20m(?) corridor around your track for moderate, to the whole slope when it's considerable. This is where you can argue that they had a reduction factor that put them at or even below 1. They were on the flattest possible route down the hill. In the picture the blue arrow shows where a local person (not part of the group) skied a steeper pitch twice before the group arrived. The red arrow indicates their route, the circle where they triggered it. Obviously they were not spaced out well enough (all 8 got caught, 4 were buried), but then again I think it is completely unrealistic to be spaced out far enough to avoid that kind of avalanche on that slope, especially as a guide with a group of seven ~ 60 year old clients. There is no way a group like that would ski the whole thing one by one, which is essentially what you would need to do to be out of harms way. My understanding is they had regrouped at a vantage point from where you can see the rest of the run (which makes sense), one person started skiing down, triggered it and it went much larger than they expected.

    I'm not trying to argue for or against Munter, I think it has a lot of benefits and some limitations. And yes, that group, if they even used it, did not do so the way it was probably meant to be used. I was mostly wondering about what seems like a very Austrian (European?) interpretation of Munter as the gold standard that is not to be questioned, thus the question about what our friends across the pond do. And yeah sorry for the confusion about 3x3 vs reduction method. 3x3 imo is a good decision aid which makes a lot of sense. The reduction method calculation is a little more debatable.

    There is also a semi-public fight going on now between the mountain guide associations and our forecasting service, because the forecasters said something to a newspaper along the lines of "I personally would not have skied this today" and "this tour was not an ideal choice for that day". One side says "we explicitly mentioned the problem in the bulletin, what more do you want?", the other side is essentially saying "this was okay by Munter, there was no way the guide could have known, how dare you imply it was a bad decision. the guy was guiding since 1974 without any accidents!". Kind of adds a new layer of unpleasantness to everything.
    Last edited by klar; 03-19-2017 at 03:22 AM.
    Ich bitte dich nur, weck mich nicht.

  13. #13
    Join Date
    Apr 2004
    Location
    Grenoble
    Posts
    343
    Quote Originally Posted by klar View Post
    There is also a semi-public fight going on now between the mountain guide associations and our forecasting service, because the forecasters said something to a newspaper along the lines of "I personally would not have skied this today" and "this tour was not an ideal choice for that day". One side says "we explicitly mentioned the problem in the bulletin, what more do you want?", the other side is essentially saying "this was okay by Munter, there was no way the guide could have known, how dare you imply it was a bad decision. the guy was guiding since 1974 without any accidents!". Kind of adds a new layer of unpleasantness to everything.
    I have to say I don't follow what happens in Austria that closely but I'd heard of these two avalanches and also the big slide at the start of the season in Austria (remote triggering) which seemed to be due to the same problem and I'm pleased you created this thread and posted some good information elsewhere (SFTE thread) as I'd not seen anything very detailed on the incidents.

    Whenever there is an incident it is almost always the case that we look at the bulletin and say "look all the information was there". If only it were so easy before the event. What you say about the group management is exactly what I said after the Tignes avalanche, it is not possible to have entirely safe spacing on that type of slope with a guided group with the leader keeping some kind of control.

    I might have skied that slope after reading the bulletin. It is true that risk 2 I would have had my guard down a bit. Even the statistical methods don't guarantee zero accidents so maybe the group just got really unlucky. In a winter with difficult conditions guides have to find snow where they can and there is going to be an element of risk taking. In Tignes the group was led by a very experienced local ski instructor although in more marginal circumstances than the one we are talking about.

    Sorry to have diverted this thread from your question to the North Americans. Hopefully they can add something about how they handle these situations.

  14. #14
    Join Date
    Oct 2005
    Location
    Sandy
    Posts
    4,681
    Quote Originally Posted by klar View Post
    I was mostly wondering about what seems like a very Austrian (European?) interpretation of Munter as the gold standard that is not to be questioned, thus the question about what our friends across the pond do. And yeah sorry for the confusion about 3x3 vs reduction method. 3x3 imo is a good decision aid which makes a lot of sense. The reduction method calculation is a little more debatable.
    Across the pond? The land of guns and heart disease? Not sure if we're the best place to look for risk reduction....
    I've skied with Canadians before that were really into the Avaluator. In the US you see the avalanche hazard triangle thrown around a bit.

    The UAC has a nice graph of risk vs. danger rating and hazard ratings over here:
    https://utahavalanchecenter.org/avalanche-danger-scale

    Nice tidbit:
    "Since avalanche danger depends on BOTH probability and consequences, different avalanche problems with the same danger rating can have very different characteristics. Some are high probability - low consequences. Some are low probability - high consequence. Some are in the middle. The avalanche problems listed here are typical of where we see them on a probability - consequence plot"

    High risk - low probability..

    Life is a lot like climbing: there isn't anything much more comforting than a good #2.

  15. #15
    Join Date
    May 2011
    Location
    Sonoma & Truckee
    Posts
    7,826
    I've only heard of Munter in passing and pretty sure I can say his methods are not widely adopted in the US. Wasn't mentioned at all during my AIARE 1 course and never seems to come up in our local snow pack discussions. The whole avalanche problem seems way too complex to try to use any kind of safety formula, as the reality is there are dozens upon dozens of variables on any give day or slope.

  16. #16
    Join Date
    May 2007
    Location
    Juxtaposition
    Posts
    5,818
    Quote Originally Posted by klar View Post
    The second, smaller slide also was not a wind slab. First day with an isothermal pack, the layer it slid on was the same kind of old facets as with the big one.
    It is too easy for routine-driven regulars to get badly caught out by a season which is different. I saw it in another country this season, guide and his team caught out badly by an irregular snowpack.

    As a more variable climate delivers a wider array of avalanche problems, it is probably good to assume even less about 'normal' and spend more time being worried about things that are different to normal and what that means for you that day. Learn to look for uncertainty and miss out on skiing what 'normally' would be ok (and probably was). This with a backdrop of some places having shitty seasons, so competition picks up.

    I'm very lucky and get to ski in and around a lot of avalanche terrain in four very different snow climates between 3 countries every year. It is one of the things that lets me see how clueless I can be. It scares me how vulnerable I'd be to the "unfamiliar" if I only skied in one place for year after year. No thanks.

    Routine sucks. Familiarity + scarcity will the be heuristics to look out for in coming years.
    Life is not lift served.

  17. #17
    Join Date
    May 2007
    Location
    Juxtaposition
    Posts
    5,818
    Quote Originally Posted by TahoeJ View Post
    the reality is there are dozens upon dozens of variables on any give day or slope.
    Analyze skier triggered incidents in the last 2 decades in your snowpack and see if you can narrow down the "dozens upon dozens" to at most one dozen weather, terrain and snowpack factors which are common to a significant number of those incidents. I bet you can. That is where rule-based decision making comes from. The trap is that there are times and places where exceptions are required. Those exceptions may merit less or more risk than the rule system indicates is appropriate.
    Life is not lift served.

  18. #18
    Join Date
    Aug 2016
    Posts
    639
    Not sure how many English language descriptions of the Munter method there are period - the only one I've encountered was published in this book:
    https://www.amazon.com/Powderguide-M.../dp/0972482733 there was talk of Munter's book being translated into English but I'm not sure it ever was.

    there are some cultural reasons it faces problems in the US 1) our lockstep individuality doesn't like hard rules and we need the conceit of processing lots of information 2) it doesn't provide great utility for the perceived type of skiing/skiing done by many current bc skiers.

  19. #19
    Join Date
    Jan 2006
    Location
    Vanity Fair
    Posts
    2,621
    Quote Originally Posted by neck beard View Post
    It is too easy for routine-driven regulars to get badly caught out by a season which is different. I saw it in another country this season, guide and his team caught out badly by an irregular snowpack.
    ...
    Routine sucks. Familiarity + scarcity will the be heuristics to look out for in coming years.
    Yes, good point and I think this is the core issue we have been having. We are on our third winter with a sketchy persistent weak layer issue and it is still catching people off guard. More than anything else this shows how unfamiliar most people are with this type of problem. Or, on the other side of the coin, how ingrained a certain understanding of what moderate/considerable means is. I think that latter issue is at least partly due to the role of the reduction method as a core strategy of risk management, which has lead to a very pronounced focus on danger level (rating) rather than danger type (wind slab, pws, etc).

    In the first season with this problem my ski partners and I had an incident that wasn't even that much of a close call (buddy remote triggered something, no one caught or in any serious danger of being caught) but really drove home how far away from our version of "normal" that was. Without the benefit of that free lesson I would be a lot less aware of just how unmanageable the issue is and a lot more likely to let my guard down on a day with a moderate rating. I still get tempted but keep coming back to that one Holy Shit moment.

    Quote Originally Posted by davidof View Post
    I have to say I don't follow what happens in Austria that closely but I'd heard of these two avalanches and also the big slide at the start of the season in Austria (remote triggering) which seemed to be due to the same problem
    Which big slide at the start of the season do you mean? Most serious incidents this season and also in the last two winters were due to the same problem and it seemed like an unusually high number of guided groups was affected, though I don't have any really solid numbers on that. The drama now happening between the guide's association and avalanche service is unfortunate and unhelpful imo. Were there any kind of legal consequences for the ski school after the Tignes slide? I assume that was taken to court in some form or other?

    Quote Originally Posted by BCMountainHound View Post
    In a similar vein as the Avaluator mention by Neckbeard, BC Search & Rescue Association has developed a protocol called RADeMS to assist with risk mitigation for tasks:
    http://www.bcsara.com/projects/radems/
    Both the German and the Austrian alpine clubs (huge organizations with a lot of lobbying power, who are also probably by far the largest providers of mountain related educational programs for recreational users) teach something similar to the Avaluator and RADeMS. The Austrian one is called Stop or Go and the German one Snowcard. These are mainly simplified versions of Munter's 3x3. None of these are quite as (weirdly?) clear cut in terms of giving you a seemingly definitive answer as the reduction method. The main effect of the Snowcard and Stop or Go is that they will keep you off slopes of a certain angle, depending on the danger level. It looks like the most notable differences to the Avaluator are that persistent weak layers are not specifically mentioned anywhere as a special problem to watch out for and that there is what seems like significantly more emphasis on slope angle. "Forest density" is also not a factor.
    Ich bitte dich nur, weck mich nicht.

  20. #20
    Join Date
    Jan 2006
    Location
    Vanity Fair
    Posts
    2,621
    Quote Originally Posted by dunfree View Post
    Not sure how many English language descriptions of the Munter method there are period - the only one I've encountered was published in this book:
    https://www.amazon.com/Powderguide-M.../dp/0972482733 there was talk of Munter's book being translated into English but I'm not sure it ever was.

    there are some cultural reasons it faces problems in the US 1) our lockstep individuality doesn't like hard rules and we need the conceit of processing lots of information 2) it doesn't provide great utility for the perceived type of skiing/skiing done by many current bc skiers.
    Interesting take on the cultural issues. I suspect another issue is that at least in the past for many regions of the US the forecasts were not regionally detailed enough to apply Munter's stuff. That book was and still is very successful in german but the english version did not sell at all and I know the guys responsible never really understood why. There have been several reprints of newer, updated versions in German (the current version includes a weather chapter I wrote).
    Ich bitte dich nur, weck mich nicht.

  21. #21
    Join Date
    Aug 2016
    Posts
    639
    That's still true somewhat. Not sure how much touring as a % of trips is done outside of areas with decent forecasts in the US though. Another issue would be whatever modifications would need to be done to the numbers within the method to fit various US regions or the US as a whole.

    Regarding Powderguide in the US there was an existing progression of books - "snow sense" "staying alive in avalanche terrain" & "the avalanche handbook" in order of complexity/technicality that were all published by the same house I think (Mountaineer's Books- available in pretty much every climbing/bc shop) and were already recommended by most curricula. At the time it was released here in the US I remember seeing it in exactly one shop - Powell's Books, the largest shop in the US. Those other 3 were everywhere.
    Last edited by dunfree ; 03-19-2017 at 06:25 PM.

  22. #22
    Join Date
    Apr 2005
    Location
    A LSD Steakhouse somewhere in the Wasatch
    Posts
    9,655
    1st ive heard of munter
    after watching someone go for a nasty ride on a stable midwinter snowqpack on a low danger day midsta low danger week
    i realized mother nature dont care what risk migration method you use
    she can send a black swan to poop on your head at any time
    "When the child was a child it waited patiently for the first snow and it still does"- Van "The Man" Morrison
    "I find I have already had my reward, in the doing of the thing" - Buzz Holmstrom
    "THIS IS WHAT WE DO"-AML -
    ski on in eternal peace

  23. #23
    Join Date
    Jan 2006
    Location
    Vanity Fair
    Posts
    2,621
    Okay, so I have more questions about the Avaluator. Our methods essentially come down to a danger level / slope angle relation (if danger level > x, stay below angle y). There are other components but that is the core of it.

    The Avaluator seems to establish a much more general relation/matrix for snow conditions / terrain. The danger level and slope angle are components of the snow conditions and terrain block, respectively, and not weighted stronger than any of the other factors in these blocks (slope angle above or below 30 is as much of a factor as forest density if I am interpreting it correctly).

    Where does that difference come from? Is this some inherent, "real" difference between the Alps and North America (e.g. slope angle less of a factor because of different snow pack, different terrain, whatever), or does this come from the methods used to analyze the data sets or the quality or resolution of the available data (e.g. slope angle less of a factor because slope angle not known or not archived at same resolution, danger level less of a factor because lower spatial resolution of forecasts, differences in danger scale, ...) ?
    Ich bitte dich nur, weck mich nicht.

  24. #24
    Join Date
    Oct 2005
    Location
    Sandy
    Posts
    4,681
    Here's a blurb from TAR on Munter and his scoring. He is trying to adjust deaths/ski touring days, not drop the # of deaths to zero. If the alpine clubs don't like the RM score they could just move the green area to a lower RM score...

    http://www.americanavalancheassociat...April_2013.pdf

    In the book 3x3 Lawinen (Munter, 2003) and in other
    forums, the author discusses the case fatality rates of
    winter mountain skiing. An estimate of the ski tour
    case fatality rate (avalanche accidents) in Switzerland
    in the 1980s corresponds to about one death in 36,000
    ski touring days. A high number of tours per winter
    (i.e., exposure) with this case fatality rate could easily
    enter into the unacceptable region if one would use
    the HSE Tolerable Risk (TR) framework for annual
    fatality rates. A use of 1/100,000 as a base rate for
    winter mountaineering seems nevertheless reasonable.
    Compared to other risks, the ratio is rather high, but
    it could be considered the price we must pay for the
    freedom of the mountains (Munter, 2008).
    Legal cases concerning risk and negligence are often
    complicated, and outcomes can be unpredictable. If
    we want to prevent arbitrary judgments in court, it
    is important to define reasonable risk thresholds in
    winter mountaineering.
    It is possible to apply these thresholds to the
    framework of the Reduction Method (RM). This
    method is based on the assessment of five key
    variables: general danger level, slope inclination,
    slope aspect, previous skiing, and load – which are
    weighted and integrated. In short, the weighted
    general danger rating is divided by the product of
    at least three weighted observations from different
    levels: regional, local, and slope (on-site level).
    The risk level is expressed as an RM value, which in
    principle can be any number from 0 to 32. Analyses of
    Swiss accident data from the 1980s give an RM value of
    2.2 to the accident rate of this particular period, while
    an RM of 1 corresponds to the suggested acceptable
    case fatality rate of 1/100,000.
    The term “Limits” was introduced by Munter to define
    a maximum reasonable risk level, akin to “The Stupid
    Line” used by Tremper. This corresponds to a RM level
    of 2, or a fatality rate of 1 in 50,000 ski tours – close to
    the historical fatality rate from the 1980s. RM=4 stands
    for the average residual risk taken in multiple fatality
    accidents in Switzerland in the 1980s with five or more
    fatalities. This is equal to a case fatality rate of 1:25,000.
    Munter suggests keeping activities to a RM value smaller
    or equal to one and to use the extended range of motion
    given by RM=2 (Limits) only in special situations under
    special circumstances. For novice users, the elementary
    reduction method targets RM=0.5 to allow for extended
    error tolerance. However it has to be understood that
    these residual risk values always represent a mean value
    due to uncertainty in determining the input variables
    of the reduction method. For RM=1, the case fatality
    rate in a single event may have a stray effect between
    1:50,000 and 1:200,000 which is equal to a factor two
    error. Higher error factors are unlikely.
    With this approach it is possible to conveniently
    visualize the accident probabilities for different
    categories of mountain activities. Table 1 shows the
    probability of a fatal accident during the period in
    life in which they are pursuing their activity, when
    estimates of typical exposures are assumed.
    People naturally strive to get the most out of their
    chosen activities without subjecting themselves to
    unacceptable risks and a likely early death.
    A long-term study over 5000 guided touring days
    of the DAV Summit Club (source: Peter Geyer) shows
    retrospectively that the mean risk of all activities
    when keeping RM <=1 corresponds to RM=0.8. A
    reduction to the risk profile to RM 1 for most users
    therefore seems feasible and an acceptable restriction
    of freedom, when considering the benefit of a longer
    life as a ski tourer.
    This corresponds to the green area of the curve graph
    in Figure 1, suggested by Munter on the relationship
    between fatality rate and the percentage of “no go”
    situations. Further reduction of the case fatality
    rate is possible, but only at the cost of an increasing
    number of missed tour opportunities. The percentage
    of backcountry users who are willing to comply to
    the proposed rules of behavior would probably also
    decrease markedly.
    Last edited by sfotex; 03-20-2017 at 03:56 PM.
    Life is a lot like climbing: there isn't anything much more comforting than a good #2.

  25. #25
    Join Date
    May 2007
    Location
    Juxtaposition
    Posts
    5,818
    Quote Originally Posted by klar View Post
    (slope angle above or below 30 is as much of a factor as forest density if I am interpreting it correctly).
    Slope angle is weighted more than any other factor at 35 degrees or above.

    However yes, ceteris paribus, a well anchored slope will offset 1:1 a 32 degree slope. As it should. But as you can see, numerous other terrain and snowpack factors can swing that balance.

    For your research, I recommend you buy an Avaluator online from MEC. It contains 2 tools.

    1. The Trip Planner: Big picture terrain selection, hazard v.s. terrain style. There is more in this tool than you realize. It will trigger another round of research into ATES.

    2. Slope Evaluation Card: slope-scale decision making, where slope angle and hazard are just two factors. It allows for positives to offsett negatives, or negatives to offset positives.

    Keep in mind these are entry-level tools to remind inexperienced people of what should be naturally churning through their head while trip planning, and then all day on the snow.

    Regarding data resolution, Canada has a good data set. Buy this https://www.avalancheassociation.ca/...spx?ID=5143665 Volume 4 covers the preceeding 10 years. Each incident analysis is summarized in the Avaluator framework.
    Life is not lift served.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •