Results of the Cowboy experiments on the decoupling of underground nuclear explosions. These experiments show that the principle of decoupling by means of a large hole is correct. The decoupling factor of 300 is still considered the best estimate for a hole in salt relative to a tamped shot in Nevada tuff. The most surprising result of Cowboy is that decoupling is not an all-or-none effect. If the elastic limit of the medium is exceeded, the decoupling factor is reduced, but only gradually. As a result it appears possible to explode a 20 KT device in a cavity no bigger than some which already exist, without producing a signal that can be detected by the Geneva system.
Latter, Albert L., Concealment of Underground Explosions. Santa Monica, CA: RAND Corporation, 1960. https://www.rand.org/pubs/research_memoranda/RM2562.html. Also available in print form.
Latter, Albert L., Concealment of Underground Explosions, Santa Monica, Calif.: RAND Corporation, RM-2562-AEC, 1960. As of July 28, 2021: https://www.rand.org/pubs/research_memoranda/RM2562.html