I think the best thing for us to get out of shelly's accident is to check ourselves and for our partners to do the same. two or more people are more likely to notice a mistake than just one person by themselves.

The circumference of an 18" diameter tree is 56.55 inches. Why would it not be possible for a person with a 63" wingspan to reach around such a tree?

Because the bones in the arm are rigid and can't form a perfect circle of circumference 63"!

Edited to add: Sorry I see this was already answered.

Maybe I have missed this already but is it not possible to walk behind this tree? Everyone seems to be assuming she was tying the hitch blindly while reaching around the tree. If she could barely reach around the tree, would she not girth hitch the slings first and then attempt to pass them around the tree. It is a lot easier to grab the end of sling if you can barely reach around than to attempt to girth hitch them together at your maximum reach!

...If the user is not aware, not paying attention, hurrying, concerned about the next lead, whatever, then it is easy to see the possible mistake of missing the loop but thinking it was through.
http://www.ualberta.ca/...girth/girthfail2.gif

When the red sling is pulled tight it is easy to see, in probably something approaching 75% of the time, (this value is made up based on my simple tests but feel free to test it yourself) how the knot could easily have locked.
http://www.ualberta.ca/...girth/girthfail3.gif

Another idea: This was Shelly's third or fourth rappel station of the day. While she was breaking down the previous station, she only partially undid a girth-hitch holding the two slings together.......in essence creating the jammed knot scenerio. When she went to create the new rappel station, she saw the slings were joined (somehow) and assumed they were still girth-hitched together. She threw the two sling combo around the tree and attached the rope.

Would it have made more sense for her to have girth hitched the slings, and then tossed one end around the tree?

...other, she passed one sling through the second one--but not then back through itself. If one of the water knots was in the back of the tree, where the two slings met, the water knot could provide the illusion that the slings were truly girth-hitched...

I am guessing that she meant to tie the two one-inch loops of webbing together with a girth-hitch, but didn't actually girth hitch the webbing. I suspect that she may have reached around the tree and passed one sling through the other sling and then pulled the slings through until she felt something become tight.

This was my point. Surely you would girth hitch the two slings before you attempt to put them around the tree. It is unlikely in my opinion she attempted to tie this knot unsighted due to the dimensions of the tree and the impracticalities of doing it.

In my opinion it is improbable that she tied this girth hitch unsighted which also reduces the probability of not tying it correctly. Given what was found (or at least what was posted on this forum), the most likely cause was indeed incorrectly girth hitching two slings together. However, I cannot imagine the events that caused this girth hitch to be incorrectly tied.

Agreed, the previous theories just don't make sense...

To throw another idea out.

Was this their last rap of the day?

I am wonder if she was planning on leaving the webbing and biner behind. Maybe she had the slings around a horn or a broken branch with the hopes of flicking it off. The webbing could of rolled off a horn. She may not have used the tree at all. Just some other thoughts.

However, I cannot imagine the events that caused this girth hitch to be incorrectly tied.

Question for Shelley's regular climbing partners: Did Shelley like to use bowlines? Did she tie bowlines one-handed, which tends to result in shorter tails? Was she in the habit of not tying backup knots on the tail?

A 10 foot sling (measured as a loop, end to end) would go completely around the large 5 foot circumference pine at East of Eden and leave plenty of length to tie a bowline. The bowline would be tied treating the flattened loop as a single strand. I am not talking about a double bowline or a conventional bowline on a bight, although this bowline technically would be on a bight.

HYPOTHESIS:

She could have tied each sling as a bowline, leaving a loop at the load end to clip into. She could have tied two slings separately this way for redundancy, and clipped both load-end loops. This rig would not allow the anchor biner to go below the edge, but she was descending to the climb on a single, fixed strand so rope rubbing over the edge was not an issue. Even under the girth hitch theory, two 10 foot slings girth hitched together around a 20" diameter tree would not have reached the edge 15-20 feet away anyway--it would only have reached about 8 feet from the tree. This bowline rig had the benefit of allowing her to approach the edge safely on rappel beginning close to the tree.

FAILURE MODE:

If not dressed tightly and with a solid backup knot, a bowline tied in webbing can collapse and slip completely apart. (A bowline in rope is much more resistant to collapsing).

In this failure scenario, each sling would be left in a separate loop, with no knots other than the water knot joint, and each sling would be clipped to the anchor biner, consistent with the findings.

REFERENCES:

Josh "goodguy" stated from firsthand observation the pine is 15-20 feet back from the edge.

Dania "leo4aclimb" stated from firsthand observation the tree was about 20" in diameter, 5 feet in circumference.

Other witnesses have settled that the anchor biner was clipped to each of two, separate, 10' webbing loops, and the loops had no knots other than the joining water knot and had no signs of damage.