Crevasse Falls

[JasonG]

http://gearjunkie.com/crazy-crevasse-falling-video

 

Yowza.

[Fairweather]

Pretty terrifying. A useful and sobering video/post. But I wonder if it's a completely legit test. The length of rope between faller and arrester is pretty short and, hence, there isn't much dynamism in the system. I've held several "punch thorough" type crevasse falls standing up--with 37-50 meters of single dynamic rope. But I'm still not sure if these anecdotes are completely useful either either. Also, in the video the arrester is facing toward the faller in almost every example--a pose that requires him to twist around pre-arrest and isn't how it usually goes down. Still, getting drug over the lip would be absolutely terrifying and I think about this often when I'm on a glacier with only one other partner. Thanks for the post.

[JasonG]

I agree that the spacing is closer than most of us North Americans use, but these folks knew the fall was coming so they were somewhat ready. Although, they tested the downhill pull, which is worst case.

 

However, I would think that in most cases, the lead person is the one that is going to go in, and the arrestor would be facing them. But, I have never been involved in a full crevasse fall (roped, one time I fell all the way in unroped - D'oh!), just the occasional leg punch.

 

Two people roped together on a tricky glacier is certainly cutting the margin thin, unless there are other teams nearby.

[mountainsandsound]

Cool post. Thanks. I wonder if your life would be flashing before your eyes in those couple seconds it takes for your partner to drag you over the lip and into the icy chasm of death...

 

I generally try to stick with 3 for a rope team on big, heavily crevassed glaciers with hidden snow bridges. But then again I don't do anything too steep and technical. Anyone use the butterfly knots as stopper knots for a two person team? Apparently 3 or 4 tied between a team of two adds a good amount of friction during a fall.

[kurthicks]

I always, yes always, tie butterfly knots when I travel in a team of two. In the Cascades, I go with 50' out between us with kiwi coils (or a stowed mountaineer's coil of the extra rope) on both ends of the rope. Using more rope is common in Alaska (wider cracks than we usually have in the Cascades), but I have never seen any legitimate study reinforcing the effectiveness of using a (longer) dynamic rope in reducing force in a crevasse fall. In fact, Rigging for Rescue has shown that static or semi-static ropes are potentially better for use in pure glacier travel since they stretch less (which is fine given the low force of these falls) and that low stretch makes hauling significantly easier.

 

The butterfly knots work incredibly well if there is snow on the surface of the glacier (which, after all, is why you're roped up because if it was blue ice using a rope won't help either of you). If you're long-roping (say 30' or more) on a blue ice glacier, you're doing something wrong.

 

Finally, effectively managing the slack via good communication & pacing, traveling when the glacier surface is still frozen, and doing proper route-finding will eliminate almost all of the possibility of a major crevasse fall.

[JBo6]

Kurt what do you do if you come across a section you'd like to belay? Does that happen often?

[kurthicks]

There are certainly places where one would want a belay. Honestly, most of the time a very tight rope and an axe in arrest position is all I do (remember those knots I mentioned?). Otherwise, create a solid anchor (say a T-slot) and belay as per normal for the leader (and possibly the follower depending on what the leader sees when they cross the bridge). These belays don't seem to happen very often, perhaps most often near bergschrunds or over mid-season obviously sagging bridges.

 

All of the concern about a major, above-the-head, crevasse fall is quite overblown in our region in my opinion and experience. Others folks might have a different opinion and that's just fine by me.

[Fairweather]

I've always wanted to see a video of a shock-loaded boot-axe belay. I still see climbers using them over sagging bridges from time. My guess: fail.

[kurthicks]

You hit the nail on the head—avoid shockloading the system at all costs!

[genepires]

that needs to be shown at all organized club basic climbing classes. and whenever someone posts a question about how to setup a two person glacier team.

 

The real sobering fact is that they knew it was coming and still had failures. imagine the "tired zombie walk state" that is typical in a glacier plod.

 

I used to do this to my students but with different backup systems. I think theirs offers a more realistic failure than mine did. I would be too scared to make the student fall that far before catching. Anyways, we did have a few self belay failures but for the most part they nailed it well. (no butterflies, 45 feet apart)

Edited December 6, 2013 by genepires

[luvshaker]

That video was fantastic. A phenomenon this video captures well is the reality of the force applied at mid-section of the climber at their harness. As mentioned earlier, even when the climber expected the fall they could not get down immediately into self-arrest. Most people practice self-arrest without that force on their harness. The speed and distance of the falling climber increase dramatically when you add the awkwardness of 200+ lbs pulling suddenly at your waist.

 

The snow conditions were also interesting, it got a little "packed out" from the multiple trips, but it looked pretty powdery.

 

Edit. I thought they weren't wearing crampons at first.....yikes, they were.

Edited December 6, 2013 by luvshaker

[tvashtarkatena]

Knowing and watching for crevasse surface indicators, avoiding terrain more likely to be cracked underneath, generous end runs, and having an ax and a ski pole for probing keeps the view from going blue. For those rare instances that call for a little extra, a quick drop to a sitting hip will stop a bus, or at least a short one.

 

If self arrest is gonna be tough, maximizing force on the arrester by using a static line is the last thing you'd want, no?

 

Tiblocs are a nice alternative to prusiks. Quicker to rig and climb with, easier to take on/off when passing the lip - I've found they still climb fine on a 7.8mm. I don't travel with pre-rigged prusiks - more bullshit to rig and unrig before heading out, they're body weight rated only, doesn't make sense for a middle person, and ascending a separate rescue coil may be required, anyway.

 

Having a biner handy to clip your pack onto the rescue line below you before completing the ascent is another good idea. A free hanging ascent is hard enough with no pack on. Finally, I find a Texas rig - (inverted Y with footloops shorty for the chest) a lot easier for free hanging ascents.

 

Of all the foot and body punching through I've seen - every one was because the punch-througher wasn't paying enough attention. Alaska's a whole nother story, though. They put huge hidden crevasses in some really weird places there.

Edited December 6, 2013 by tvashtarkatena

[chris_erickson]

If self arrest is gonna be tough, maximizing force on the arrester by using a static line is the last thing you'd want, no?

 

Great video, however to me the rope spacing is the main issue. I use low-stretch rope exclusively in glaciated terrian (barring an approach to a technical climb) and also train regularly with it. Even novices are able to catch a real fall, downhill, while not looking at the faller if the rope spacing is adequate for them.

 

That said, it's a trade off. Dynamic rope = more stretch, worse cut in to the lip, longer to haul/ascend, greater likelihood of catching a leg/crampon during the fall. Low stretch rope = less stretch, higher force to arrester, less cut in to lip, less distance in a fall.

 

The Denali Ranger/Guide community is a bit split on the subject. Rangers and some guide services use low-strech. Other guides use dynamic.

 

I also use knots in the rope, but perhaps in more specific terrain such as: steep (difficult to arrest), low-friction snow, inexperienced partner, shorter rope, skinning (i.e. ice ax not in hand) and others. Kurt's probably smarter just to use knots all the time.

 

Last thing - I train with real crevasse falls quite often as in the video. Personally, I tie the backup safety line to the faller, not the arrester. In many of the video clips, the arrester's ice ax is quite close to the rope coming under tension. Low likelihood event for sure, but I feel better having the backup line completely out of the way.

Edited December 6, 2013 by chris_erickson

[tvashtarkatena]

Yeah, spacing is the most important thing for novices, all other things being equal.

 

A nit, perhaps, but a static line increases the force on the lip - and all parts of the system - so it should cut through the lip more than a dynamic. I wouldn't think the very minor change in diameter of a dynamic rope due to stretch would not be enough to overcome this factor, given two ropes of equal diameter.

Edited December 6, 2013 by tvashtarkatena

[chris_erickson]

I could be wrong here, but in my anecdotal experience it's the other way around.

 

Dynamic ropes absorb more force than low-stretch ones. So if I have it correct, Dynamic lines = less force on anchor, more force (tension) in line. Low-stretch lines = more force on anchor, less tension in line. Since the dynamic rope absorbs more force, i.e. a higher unit of tension, it cuts more.

 

The decreased diameter of the tensioned dynamic rope, again anecdotal, seems to create a substantial difference in lip cut. My falls on low-stretch ropes seem to cut a long ways, but falls on dynamic lines always cut more.

[tvashtarkatena]

With all due respect, a static line always puts more force on all parts of the system because deceleration, and therefore peak load, is greater. That's why its never a good idea to lead on a static rope, and that's why ice climbers with sketchy shorty screws love climbing on double skinnies that stretch like rubber bands, even if that means a longer length of fall.

 

Force and energy are two different things. The energy absorbed in a fall is the same for both dynamic and static ropes. The force on the system is not, because force depends on deceleration rate at which the kinetic energy of a fall is absorbed. The faster the deceleration rate (ie, the more static the rope is), the greater the peak load or force on the system.

 

This is why the old timers used dynamic belays like sitting hip etc with those old static hemp ropes - they didn't want to a) tear each other in half and b) they wanted at least a prayer that some part of that system would actually withstand the peak load.

 

If you have anecdotal experience that dynamic ropes cut more, its likely because a) they are significantly skinnier than the static line with which they're being compared or b) their stretch reduces their diameter such that its significantly smaller than the static line-thus overriding the latter's lower forces on the lip, or c) the conditions at the lip weren't the same during the test.

Edited December 6, 2013 by tvashtarkatena

[chris]

I think the dynamic vs static is determined by what the rest of the climb requires. For the occasional short rock step or bergschrund crossing I'll use a "rando" rope and double it up for the technical "pitch". Its nice to have a lighter rope for the DC, Coleman-Deming, etc.

 

But if I'm anticipating technical pitches, the dynamic comes out. The Cascades, like the Alps and western Canada, have countless examples of technical routes that require moderate glacier travel beforehand and/or afterwards.

 

I always use butterfly knots in a team of two. I've even used them on really broken terrain with a team of three. DPS caught me in a crevasse fall in the Ruth Gorge about 10 years ago using this method. And I've been able to catch every "practice" scenario in my guides' training and examination using this method too.

 

Lastly - learn how to use a garda hitch - it has a couple of great applications in crevasse rescue for ascending a rope and providing an auto-belay on the haul system.

Edited December 6, 2013 by chris

[mountainsandsound]

 

Lastly - learn how to use a garda hitch - it has a couple of great applications in crevasse rescue for ascending a rope and providing an auto-belay on the haul system.

 

Have not tried it for ascending- but have used it in practice for hauling. The friction penalty seemed high, but the self-tending ability was great. A good one to add to the toolkit I think.

 

Maybe I need to get different cord to make a prusik knot that will catch the rope better when hauling. Mine is 6mm and seems stiff as hell.

Edited December 6, 2013 by Nater

[chris_erickson]

If you have anecdotal experience that dynamic ropes cut more, its likely because a) they are significantly skinnier than the static line with which they're being compared or b) their stretch reduces their diameter such that its significantly smaller than the static line-thus overriding the latter's lower forces on the lip, or c) the conditions at the lip weren't the same during the test.

 

A and C would seem fairly obvious, don't you think?

 

When I've seen this occur, my best guess was that dynamic ropes cut in to lips more because the dynamic rope was moving (stretching) substantially more while weighted. The low-stretch events are over much quicker (assuming a solid arrest), and when the weighting of the rope occurs, the event is pretty much over. Thus the rope doesn't "saw" the lip as much. Dynamic ropes keep stretching after the arrest, and as more and more of the load transfers to the anchor/rope, it cuts more.

 

The video posted should be a pretty cool reminder to practice self arresting and practice good rope spacing. I think we may have dragged this down in to the weeds, sorry!

[Tod]

In the video, if the climbers had knots in their ropes, or a third person on the team the falls would have had a I rate of success (not more distance).

 

Knots:

-2-Man Team: Knots will most likely stop the fall, not the climber. Butterfly knots can be incredibly effective and make a two man team realistic in crevassed terrain.

-3-Man Team: Multiple team members to stop the fall adds to the success rate of catching a fall.

 

Distance between climbers:

-More distance: adds more potential slack which adds more potential force on the climber arresting the fall. (you have 100' between you and someone falls, the person falling will have picked up a lot of speed/force prior to the rope going tight)

-Less distance: adds better reaction ability prior to too much force being added to rope/climber arresting fall. (you have <50' then the rope will go tight sooner and less force will hit the person catching the fall)

-Correct Distance?: I see a lot of teams that assume to use the whole rope, which makes for far too much length between climbers. Being closer to the other climbers make for quicker actions under less force and better rope management. Obviously though you need enough length to get around crevasses and stop a fall before you get pulled in. AAI has a great page showing info on rope lengths between climbers (http://blog.alpineinstitute.com/2010/11/rope-length-on-glacier.html)

 

[Fairweather]

-Less distance: adds better reaction ability prior to too much force being added to rope/climber arresting fall. (you have <50' then the rope will go tight sooner and less force will hit the person catching the fall)

-Correct Distance?: I see a lot of teams that assume to use the whole rope, which makes for far too much length between climbers. Being closer to the other climbers make for quicker actions under less force and better rope management. Obviously though you need enough length to get around crevasses and stop a fall before you get pulled in.

 

I'm having a hard time with this. As another poster pointed out, "force depends on deceleration rate at which the kinetic energy of a fall is absorbed. The faster the deceleration rate the greater the peak load or force on the system." Let's compare this to climbing a rock wall: what you are saying, Tod, is akin to claiming it is easier to hold the fall of your follower when he is, say, 5 meters from the belay than it is when he is fifty. This simply isn't so. If you're talking exclusively about opportunities for rope slack, then your point is probably valid. But if the argument is limited to dynamic ropes--kinetic force applied as a function of deceleration over time--then it seems to me more space between climbers is better as the deceleration time is longer, hence, peak load is lower. What am I missing here?

 

 

 

Edited December 8, 2013 by Fairweather

[ScaredSilly]

Having a biner handy to clip your pack onto the rescue line below you before completing the ascent is another good idea. A free hanging ascent is hard enough with no pack on.

 

Pre-rig it. Much better than screwing around with having to do it once yer in the hole. The same for prusiks or tribloc.

[glassgowkiss]

You hit the nail on the head—avoid shockloading the system at all costs!

I don't think it's possible over a crevasse. Most likely if the bridge collapses climber will be in the middle, so automatically there is going to be normal leader fall situation.

[glassgowkiss]

Also me thinks there is no reason to use rope while on skis, maybe on uphills, but I think on skis there is no way of really arresting a crevasse fall. The vid made me re-think the whole glacier travel issue.

[pcg]

Also me thinks there is no reason to use rope while on skis, maybe on uphills, but I think on skis there is no way of really arresting a crevasse fall. The vid made me re-think the whole glacier travel issue.

On loose snow conditions as in the video it may not be possible to self arrest, but because one is on skis does not prevent you from dropping down and planting an axe. Butterfly knots in the rope might help as well. It does make you rethink the whole roped up/suicide pact scenario though.