U.S. avalanche fatality trend is flat for the past 22 seasons

Where were you during the winter of 1994/95?  Knox Williams, Bruce Tremper, and Mark Moore headed up the three big avalanche centers in Denver, Utah, and Seattle.  Jim Kanzler was running the center in Jackson Hole, Brad Ray led the avalanche center at Mount Washington, and a small avalanche center I started in Bozeman was in its fifth season.  Avalanche centers didn’t even exist in the Chugach, at Tahoe, or Mt Shasta.  Almost all backcountry skiers used telemark gear, with some (including yours truly) in leather boots.  Snowmobiles didn’t come close to the machines of today in terms of power, mobility, and weight.  Most ski areas had closed boundaries and the term “sidecountry” wouldn’t be widely used for more than a decade.  Finally, the Internet was only just starting to gain traction, with the first avalanche advisories being posted to the web at the end of that season.

Would it surprise you if I told you that since that 1994/95 winter the annual number of U.S. avalanche fatalities has not increased?  That’s right, for the past 22 seasons as we’ve transitioned to smart phones, social media, lightweight AT gear, powerful snowmobiles, and a literal explosion in backcountry use, the number of seasonal U.S. avalanche fatalities have remained steady (Figure 1).  Also interesting is that the number of snowmobile fatalities is flat over this time period (Figure 2), while there is some evidence that the number of backcountry skier/snowboarder fatalities may be rising slightly (Figure 3). 

Figure 1: U.S. avalanche fatalities from the 1994/95 winter through the 2015/16 winter.  The slightly decreasing least squares trend line is not statistically significant (p = 0.7), indicating that there is no statistical evidence of a change in the number of avalanche fatalities during this time period.

Figure 1: U.S. avalanche fatalities from the 1994/95 winter through the 2015/16 winter.  The slightly decreasing least squares trend line is not statistically significant (p = 0.7), indicating that there is no statistical evidence of a change in the number of avalanche fatalities during this time period.

Figure 2: The number of snowmobile avalanche fatalities, like the overall fatality rate, has not changed over the past 22 seasons.  The slightly decreasing trend line is not statistically significant (p = 0.6).

Figure 2: The number of snowmobile avalanche fatalities, like the overall fatality rate, has not changed over the past 22 seasons.  The slightly decreasing trend line is not statistically significant (p = 0.6).

Figure 3: The number of backcountry skier and snowboarder avalanche fatalities is also relatively flat, though there is some statistical evidence (p = 0.07) of a slight increase over the time period.

Figure 3: The number of backcountry skier and snowboarder avalanche fatalities is also relatively flat, though there is some statistical evidence (p = 0.07) of a slight increase over the time period.

This flat line would be no big deal if backcountry use was also flat.  However, anyone who has been in the backcountry for the past 22 years knows full well that – anecdotally of course – use has skyrocketed.  Although it is challenging to get good, solid numbers on dispersed winter recreation, we can utilize avalanche advisory usage as an imperfect proxy.  For example, over the past 22 seasons the Utah Avalanche Center reports a 12-fold increase in avalanche advisory usage, the Colorado Avalanche Information Center’s usage increased 17-fold, the Gallatin National Forest Avalanche Center has seen a 36-fold increase, and the usage at the Northwest Avalanche Center increased 60-fold!  Clearly, some of these increases can be attributed to the ease with which folks can obtain avalanche information these days, but this still indicates many more people are going into the backcountry.

Assuming a conservative estimate of use increasing 8 times and combining it with our flat fatality trend means our fatality rate (avalanche fatalities per backcountry user day) has dropped dramatically. In fact, this suggests that our fatality rate has dropped by at least a factor of 8 (and probably more) over the past 22 years.  If our fatality rate had stayed steady while the use increased we might well expect over 200 U.S. avalanche fatalities per winter!

Our community will continue to do all we can to push the number of fatalities toward zero.  However, we also need to take a step back and recognize that this flat line for U.S. avalanche fatalities is big deal.  It’s a win for avalanche educators at all levels, from those providing professional courses to those giving Know Before You Go awareness presentations.  It’s a win for backcountry guiding and ski area operations that work to protect and educate their clients.  It’s a win for equipment manufacturers who have developed an array of great equipment, including much improved avalanche beacons, Avalungs, lightweight helmets, and airbag packs.  And, it’s a huge win for our regional avalanche center network that provides the public both with avalanche education and with current, solid avalanche information for the areas where folks are recreating.  Ultimately, this flat fatality trend during a period of explosive backcountry growth shows that what we are doing works, and that is something that should make all of us proud.

Karl Birkeland

Acknowledgements: The Colorado Avalanche Information Center provided the data for this short post.  Our entire community owes a huge debt of gratitude to the CAIC, the CAIC personnel that compile the data, and the other folks who report and document avalanche accidents so that the rest of us can benefit from the lessons learned.  Thanks also to Simon Trautman, Doug Chabot, and Spencer Logan for their constructive reviews and feedback. 

Posted on June 27, 2016 .