I really enjoyed reading your post. Glad that I stumbled across your blog!
David, thanks for another of your careful insightful summaries of helpful information about skiing. You are amazingly good at it.
I can’t say I would have predicted those results. The opposing argument is that as safety equipment and information improves, humans use it to go closer to the edge in raising the level of risk. But here I think that more people are getting out more often, but the overall burial rate is not rising.
I have imagined that the increased reliance on avalanche beacons leads to more people taking stupid risks with the justification that “it will be OK because I’m wearing a beacon”. But perhaps it’s more like that even if beacons were not sold as magical safety talismans, people who are open to taking stupid risks would just find a different justification.
The more detailed inferences are tricky to assess because we’ve got much more accurate and complete info about the negative outcomes than we do about the exposure base. Like here’s two “bias” trend factors which I’d _guess_ are in the exposure base through the period of the study: party sizes are generally getting smaller, and more parties are not “guided”.
But I’m not criticizing, because I don’t see how we’re likely to get much better data about the exposure base in the future—so might as well make inferences from what we’ve got.
It’ll be interesting to see what will happen to these trends in the next 10 years, with a whole new kind of information from user-supplied trip reports (and videos).
Reading the original paper was easier than I expected—good thing you posted the link.
The number of _complete_ burials per year does not show any trend over the period of the study.
I think the number of partial burials _reported_ does show an increasing trend (Figure 1 in the paper), but the authors interpret that as a changing trend in how incidents are reported and/or recorded (and I assume they are correct).
The authors are well aware of the problems with lack of accurate knowledge about the exposure base.
Their analysis of low versus moderate versus considerable risk levels was an interesting attempt, but I think it’s a very tricky question and I didn’t find it convincing. I think a more careful statement would be, “No definite evidence was found of some trend from 1988 to 2006 that recreationists are taking on more or less avalanche risk—but the exposure base and definitions are so sketchy that who knows.”
Actually I think the _overall_ data of the paper supports the idea that recreationists are taking on _less_ avalanche risk on a trend over the longer period from 1977 to 2006. Simply . . .
More people per year are getting out there, but only the same number per year are getting completely buried.
This is independent of people getting better at using avalanche beacons, and independent of tricky questions of low versus moderate versus considerable.
It rests on three assumptions:
(1) number of complete burials is reported fairly accurately.
(2) true number overall serious avalanche accidents correlates with number of complete burials - (though there are several other ways than burial to die or get seriously injured from an avalanche)
(3) “true avalanche risk probability” correlates with true number of serious avalanche accidents - (the other possibility is that there’s a trend of recreationists taking on more “true avalanche risk probability”, but just getting very lucky in not suffering the consequences).
Actually the “true probability” for a specific ski tour on a given day is unknowable, but surely depends substantially on several factors other than the official risk level 1-5 number for the day.
I think those assumptions are reasonable for the data used in the paper.
Thanks for those thoughtful comments Ken.
> I can’t say I would have predicted those results. The opposing argument is that as safety equipment and information improves, humans use it to go closer to the edge in raising the level of risk.
I would have assumed the same. For example some studies show that helmet wearing, seatbelts, airbags all lead to people taking more risks. I’m not sure the Swiss data can necessarily be extrapolated to other areas though. My experience of Swiss skiers is that they are more risk averse/conservative than the French - there is less focus on extreme descents etc. It would be interesting to see the French / US figures to see if there is any trend.
It is interesting that the overall burial trend has not increased in line with the increase in skier numbers. This must be down to less risk taking as you say, but why? Again researchers like McCammon claim that as skiers gain knowledge they tend to take on more risk so you would expect burials to increase at the same rate as the increase in skier/days.