Rockfall and cut slings

When travelling light, I bring thin, 1/8” (3.2mm) 12-braid UHMWPE (generic Dyneema) slings. I use “class 2" bury splices to make eyes on each end. My tests give a breaking strength of 2400 to 2700 lbs when these slings are pulled end-to-end (eye-to-eye). However, my friends often express worry that the thinness of these slings makes them susceptible to cuts “from rockfall;” they are generally more comfortable with 16mm nylon tubular webbing (2200-2400 lbs unknotted; but substantially weaker with a knot).


It is pretty hard to test the relative cut-resistance of these slings, since the rockfall scenario has so many parameters. If one drops a 500 lb rock from 20’, and it hits just right, it will sever any polymer rope, and might even sever 1/4” stainless steel cable. So the test described below is scoping in nature, and merely intends to give a feel for the relative resistance of the two sling types.  The results mainly give me hints how I would design a more meaningful test.

 

The test: I had two textured concrete (with silica sand fill) landscaping pavers gorilla-taped together, to make an 8” by 12” “anvil.” Across the 8” width, I placed either 6 pieces of 16mm nylon webbing on 16mm centers, or 6 pieces of 1/8” (3.2mm) UHMWPE on 16mm centers. From 5.0’ above this anvil, I dropped a 3.0 kg rock, 15 times for each set. (I had previously practiced so I could consistently get the rock to fall near the middle of the anvil each time.)

 

Since I wanted the slings to have full strength, I sewed eyes on the ends of the nylon slings with 540 stitches of Mara 50 thread (5.3 lb tensile strength), resulting in > 4000 lbs strength for the eye stitching. A test breaking a new piece (so sewn) gave 2334 lbs at breaking, quite close to the manufacturer’s (Bluewater) specification of 2200-2400 lbs. (The break did not occur in the sewn ends, but in the middle of the sling.) For the UHMWPE, I used class 2 bury splices on each end, with ~14 (not tight) stitches of 100 lb woven spectra fishing line to prevent the ends from accidental unburying. A test sample of new UHMWPE line gave ~2700 lbs at breaking.

 

Here's what this rock (a massive serpentine marble) looked like after the tests:

 rock1rock2

The rock started at 3.03 kg, massed 3.02 kg at the end of the 1st test (with 16mm nylon), and 2.99 kg after the 2ndtest. In the second test, I could see a few flakes of rock breaking off, making the edges sharper. You can see the powdery impact marks on the right photo.


Here’s what the slings, on the same landscaping pavers, looked like after the test:

16mmuhmwpe

 

 

The arrow points to a place where the rock hit twice; I refer to this as the “unlucky section” of UHMWPE. In one impact at this spot, a small fragment of rock broke off and stayed right at the impact site; a subsequent drop hit on top of this rock fragment.

 

After the test, each sling was pulled to breaking with a come-along, the force measured by a crane scale.

 

Here’s a plot of the results:

results

 

The UHMWPE did relatively well compared to the 16mm nylon; but none of the results are really “good.” As expected, the UHMWPE that was “double-hit” in one place broke at a rather low tension. A “surprise” was the rather low breaking strength of 16mm nylon sling 1; this sample seemed to have only a slight nick at the junction with sling 2. However, since tubular webbing has no guard thread running near the edge, it doesn’t take much to cause a catastrophic tear from one point of compromise.

 

If I were to repeat the tests, I would:

1)      Use more replicates;

2)      Drop several smaller (1 kg) rocks at once, through a 6” diameter plastic pipe, with a hopper feed;

3)      Alternate nylon and UHMWPE samples to minimize bias from the degradation of the substrate and falling rocks;

4)      Repeat this all with the anvil tilted 45 degrees to horizontal, to achieve a more “glancing” blow;

5)      Use a better load cell with higher time resolution for the break tests.

The type of rock dropped would be a mix of local limestone, basalt scoria, or sandstone chunks, with a bias to selecting more equant rocks with less dramatically sharp edges. Originally the impact surface had textured indents “ledges” running perpendicular to the webbing; I felt the texture would be more representative of harsh conditions and might cut the straps from below. I would probably still use this surface, though it does seem the worst damage was mainly on the dropped-rock impact side of the slings.

 

 

There is a concern that rockfall will cut webbing and rope used in climbing. I have seen several handlines that were apparently cut by sharp limestone rock fragments; I actually witnessed ~100 lbs of rock fragments (knocked down by another climber) fall and sever a 16mm webbing handline, whose end was about 20’ below us, on a talus slope below a cliff. I’ve also found a rope handline, severed by spring rockfall off of a limestone cliff. I understand these concerns, especially for people who climb in areas that have heavy snows, and freeze-thaw cycles. In both cases, the cuts were not near the anchor, but were below the cliff on the talus slope, where rocks could fall off the cliff for a substantial distance and inpact the "loose end" of the soft goods.

The evidence for slings cut by rockfall

First, I am a fan of redundancy in anchors, especially when it is easy to do. When there are two bolts, by all means clip each so there will still be one weight-sustaining point, if the other fails. I guess I came from a time when bolts were often untrustworthy and many anchors were on pitons. But lately, most people seem to want redundancy in the slings, not because of worry about the actual metal-rock connection, but because of the possibility that a soft-goods sling will be cut by rockfall.

There are anecdotal examples of people being saved by sling redundancy, after one leg was cut by rockfall. One of the more popular examples is given in this thread:

https://www.mountainproject.com/forum/topic/110088639/rock-fall-results-in-chopped-anchor

 

Big And Little Cottonwood Canyons are adjacent in the Wasatch (SE of Great Salt Lake), on N and S sides of a quartz monzonite ridge. "The Thumb" routes (site of the accident) comprise one set of climbs in Little Cottonwood Canyon.

Note that they were climbing on a warm February day, in an area known for severe rockfall, in times when the snow is melting above. The area gets cold at night, and frost wedging undoubtedly occurs, so rocks cut loose after the sun warms the rocks.

Note also that the anchor that they clipped had previously been smashed by rockfall.

Of 122 climbs logged in Mountain Project alone for general thumb, 88 for this variation, this is the only one to mention rockfall (as of the date of this article). The logs generally don’t mention other partners. A microwave-sized fragment of granite (my microwave is 24” by 18” by 13” without legs) would weigh 500 lbs, so the original boulder, which had at least three such fragments, may have been 2000 lbs. HOWEVER, there have been many reports of rockfall-caused deaths and accidents in the areas of Big and Little Cottonwood Canyons; the people who were killed or severely injured probably didn’t log in to report their climbs

The AAC analysis of the accident is here:

http://publications.americanalpineclub.org/articles/13201213756/Rockfall-Anchor-Chopped

Note the AAC article mentions that the hanger was smashed by this accident, but fails to mention that the hanger was already smashed some when the climbers first appoached, enough to cause them some trepidation about clipping in. I think there was a pre-bias toward showing how important it is to be redundant. Here is a screenshot of the MP thread:

mp_thread

In April, 2018 (2 years later) another climber (on thumb) had to bail when rockfall damaged his rope.

Recently, a woman (wearing a helmet) was walking to a climb in Little Cottonwood and hit by rock:

https://www.fox13now.com/news/local-news/i-just-cant-live-without-her-mother-of-hiker-severely-injured-in-little-cottonwood-canyon-speaks

it seems that she eventually died.

 

Another woman was killed by rockfall in Big Cottonwood Canyon, Utah while belaying her partner; as a rleatively minor point, the rope was cut by rock.

https://www.climbing.com/news/one-killed-another-rescued-in-utahs-big-cottonwood-canyon/

The survivor (her boyfriend, also very injured) pleads for people not to climb “early” in that area (his fateful climb seems to be in May):

https://www.youtube.com/watch?v=YzS9V_ArXfk&ab_channel=KUTV2NewsSaltLakeCity

 

Here’s the report of another climber killed by falling rock in Little Cottonwood Canyon:

https://www.sltrib.com/news/2021/10/11/rock-climber-killed-fall/

 

Climbing in the Cottonwood Canyons, especially early in the season, seems to have exceptional danger from falling rock.  Many studies, for various OTHER areas, come up somewhat differently:

 

4.5% of climbing accidents from rockfall:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6981967/

 

Broad analyses of rock-climbing fatalities

https://blog.gitnux.com/rock-climbing-death-statistics/

https://www.improve-climbing.com/how-dangerous-is-climbing/

 

falling rock cause of over 26% of climbing fatalities, 7 % injuries+ fatalities

https://www.liebertpub.com/doi/10.1089/ham.2021.0085

 

falling rock and ice 21% (of those, 56% head trauma)

https://bmjopensem.bmj.com/content/8/1/e001281

 

My general take is this. First, I very happy that the redundancy at an anchor, where the sling on one side was cut, apparently saved lives in the "Thumb incident."  But I also think they were just lucky to have escaped with “only” severe injuries. What should they have thought, after seeing that the hanger had already been smashed by rockfall, knowing that they were climbing after an early-season thaw? I would argue that under the conditions of the accident, the “expectation” of a sling getting cut by a 500 lb rock falling at least 20’, is lower than the expectation of getting killed directly by the same rockfall.