The northwest climbing community experienced a string of tragic accidents in September of 2014, when three veteran climbers were killed in accidents taking place at a bolted sport crag, a huge, scruffy alpine wall, and the steppy approach terrain below a climb. The common thread linking all three deaths was a simple yet fatal accident while rappeling. The idea of rappels being statistically the most dangerous part of a climb is drilled into new climbers from day one, and the pages of each year’s “Accidents in North American Mountaineering” are filled with examples of rappel errors leading to deaths. Unlike when going up, climbers tend to rely 100% on their gear being used and functioning correctly for safety on descents. Climbers are also more likely to be going back down while fatigued, hungry, impatient, or in darkness, relative to their situation on the ascent. Some of these variables can be controlled and others must be accepted, but there are several basic improvements, both in terms of gear used and strategies employed, which can contribute to safer rappels. To make safer rappels, we need to analyze the most common accidents causes: mistaken device rigging, miscommunication, and rappelling off the end of the rope.
Rappel/descent accidents often occur when first starting the rappel or leaning back from an anchor. This failure is caused by erroneously setting up one’s rappel, or else by believing that one is going to be lowered by a belayer, but with a belayer who thinks the climber will rappel. The simple act of having both partners clearly and repeatedly confirming a plan for how a climber is to get down a given pitch (either at a single pitch crag or coming off an alpine wall) should prevent the scenario of a climber being mistakenly taken off belay while he or she thinks that they will remain on belay and be lowered. Yet this kind of accident occurs frequently, especially at popular and crowded crags where communication is difficult and a plan wasn’t mutually agreed upon before starting. When I lead single pitch routes, I typically say “You will lower me, ok?” or “I will rappel, ok?” and then wait for a confirmation before going through the more standard string of “on belay”, “belay is on” commands. This kind of question and answer communication is just as important as the traditional belay commands, but the lack concern about a leader getting off a pitch is common. The second major way in which accidents occur while starting a rappel is via incorrectly setting up a rappel device. At rappel stations, I prefer to nearly always be hanging or partially weighting the tether which I use to clip in to an anchor, even if there is a ledge. Keeping weight on this life-line prevents another climber (or myself) from mistakenly unclipping my locker from the anchor when things are being shuffled around. For me, this tether is typically a single sling girth-hitched to my belay loop and clipped in with a small, keynose locker, which is much easier to disconnect form small chain links or taut webbing than a notched locker. The typical accident in this scenario will occur when a climber thinks that he or she has correctly set up their rappel
system, but does not fully lean back and forcefully weight/bounce their rappel device before disconnecting from the anchor. This lack of pre-testing is likely due to both a mix of assuming that one has set things correctly, and a reluctance to bounce/hang at a belay station until forced to lean back while descending. Regardless of the reason, climbers should always aggressively weight their rappel setup while their anchor tether is clearly slack, before they consider disconnecting for a rappel. The more clutter one has on their harness and belay loop, the more difficult it will be to ensure a correct setup, so strip extra gear from this area of the
harness before descending. A excellent safety measure for multi-rappel descents is to use a no-twist locking carabiner (Metolius Gatekeeper, Black Diamond Rock Lock, etc) and leave it in place locked on the belay loop for the entire descent. Another excellent option is to place your locker in the dogbone-style inner loop found on CAMP and CASSIN harnesses, which allow any carabiner to remain securely locked in what amounts to a belay loop within your belay loop. Either of these mitigation strategies will reduce the number of variables that can go wrong with a rappel’s rigging.
The second major way in which rappel accidents occur is via rappeling off of one (or both) ends of the rope. The rope may be mis-threaded through the anchor, and although actually sufficiently long, it is hanging off-center for the rappel, making one end too short. The best way to prevent this is by having one (and only one, some ropes come with additional marks a few meters from each end)middle mark or pattern change, which is confirmed by both climbers to be at the anchor before rappelling. If your rope has no middle mark, use a sharpie to put one in place. If you are on a climb and find no middle mark, or a faded mark, use a small piece of climbing tape, a band-aid, or some chalk to mark the middle. And if you are rappelling in the dark, know ahead of time how many of your arm-length pulls will constitute half of your rope. For reference, expect 18-25 full wingspan pulls will be half of a 60m or 70m cord. The other means of rappel failure is by simply sliding off the ends of a rope after centering it correctly and rigging your rappel correctly, but when the rope doesn’t reach the next anchor (or a climber looses grip on the rope). Knotted ends (individually, rather than together which induces twists) are a 100% effective method of preventing this. Yet knots bring their own set of potential problems, so many climbers don’t use them. There are a few strategies and pieces of
gear which make a climber less likely to slide past their ends, and don’t carry the potential downsides which knots do. Rigging a quick prusik or friction-hitch backupis a fast way to be able to go hands-free and stay on rappel. This method will work in conjunction with any rappel device, but works especially well if you are extending your rappel device up off your belay loop, because it put more space between your harness and your device. Another strategy, and one which I see
becoming increasingly common, is to rappel on a GriGri or one of the new locking (yet non-mechanized) devices such as a Mammut Smart or the Edelrid Mega Jul. All of these devices have the built-in feature of locking up on a rappel when the climber isn’t actually releasing them to initiate the descent. Of the three models, I prefer the Mega Jul by Edelrid, as it is the lightest, has two slots, and can be rigged (by rotating it 180 degrees) to function as a normal ATC-style rappel device, with no additional capabilities. Any of these three devices (or a friction hitch backup) can still allow the ends of a rope to pass through them if a climber isn’t careful. Yet they will allow one to stay on the rope despite getting knocked unconscious, slipping on wet rock and loosing your grip, gettingstung by a swarm of hornets, or letting go to remove a stubborn piece of gear. Using these self-locking devices, I will often slowly approach the final few meters of a rope-stretching rappel, and then pull up the tails and tie knots in the ropes, especially if the end of the rappel requires a pendulum. As the first climber down such a rappel, I’ll thread one end of the rope through the next anchor (or clove it to myself) and clove off the other end of the rope to myself as well. By doing so, I would have prevented my partner from rappelling off the rope ends, but haven’t left knots in the rope which would need to be untied.