This paper outlines transparency measures that could be taken to build international
confidence that "subcritical experiments" conducted underground by the U.S. or any
other country do not violate the Comprehensive Test Ban Treaty. Our calculations suggest
that on-site measurement of the integrated radiation output from these experiments
should be adequate to verify that nuclear yields greater than grams of TNT
equivalent had not occurred, even allowing for the possibility of substantial shielding.
However, our calculations also indicate that a pure fission experiment that remained
strictly subcritical for its entire duration should have a yield no greater than 0.1 microgram
TNT equivalent, even given irradiation by an external deuterium-tritium neutron
source. Experiments involving chemical-implosion-induced deuterium-tritium
fusion could have yields above this limit, and may require fission product measurements
in addition to prompt radiation measurements to rule out a supercritical test.
In October 1995, the Department of Energy (DOE) announced a series of
experiments which it said would help determine the impact that aging plutonium
and new methods of producing replacement plutonium components for
warheads will have on the reliability of the u.s. nuclear weapon stockpile.
This series of "subcritical experiments" is to be conducted underground at the
Nevada Test Site. In a 1996 article in The Bulletin of the Atomic Scientists, we
suggested that the need for each proposed subcritical experiment be reviewed
by an independent panel; that a study of above-ground alternatives be conducted;
and, in the event that any experiments go forward, that arrangements
be made for transparency measures sufficient to build international confidence
that the experiments are subcritical.l Subsequently, the DOE decided
that the need for the experiments had already been adequately reviewed by
the JASON group of consultants,2 but that it would support a study of
above-ground options by the JASONs.3 At the time of this writing, it has not
agreed to provide the international community with any physical data bearing
on the subcriticality of the experiments.