Washington, May 25, 1966.
THE IMPACT OF A THRESHOLD TEST BAN TREATY ON SOVIET MILITARY PROGRAMS
To estimate the impact of a Threshold Test Ban Treaty on Soviet military programs, with particular emphasis on its impact on Soviet ABM activities; to discuss the capabilities of US intelligence to monitor such a test ban; and to evaluate Soviet capabilities for covertly violating it.
The Threshold Test Ban Treaty considered in this estimate is generally in line with proposals under discussion in the US and abroad, but it does not reflect a specific proposed treaty. We assume continuation of the terms of the Partial Test Ban Treaty now in effect, which prohibits testing of nuclear devices in the atmosphere; beyond its limits, including outer space; underwater (including both territorial waters and the high seas); or in any other environment if such explosion causes radioactive debris to be present outside the territorial limits of the state under whose jurisdiction or control such explosion is conducted. The Threshold Treaty considered in this estimate would add to the restrictions of the current treaty a prohibition of any underground nuclear test producing a seismic disturbance above 4.75 on the Gutenberg-Richter scale. It is assumed that treaty language will specify a method by which a mean magnitude for any particular event will be established, thus avoiding international disputes about the magnitude of reported events. It imposes no limitations on the underground medium in which the tests take place, or on the degree of decoupling employed. It provides no on-site inspection and no sanctions. Each participating national will have to decide for itself whether any given seismic event of magnitude greater than 4.75 was caused by a nuclear detonation and constituted a treaty violation.
A. We believe that for most of the Soviet military development programs which we can foresee over the next few years a Threshold Test Ban Treaty would impose no greater restrictions than those already imposed by the Partial Test Ban. However, the relationship between the yield of underground explosions and the resulting seismic readings is uncertain at best, and can be greatly altered by decoupling. The Soviets might therefore still test over a wide range of yields, depending on how far they were willing to risk violating the treaty and to support the cost and effort of decoupling. Practically speaking, we believe that they could develop weapons yielding [less than 1 line of source text not declassified] by scaling up from lower yield tests which would have a fair chance of not producing seismic readings above 4.75. They might conceivably develop TN weapons with yields [less than 1 line of source text not declassified] by expensive decoupling methods. (Paras. 13-19)
B. With respect to ABM weapons, we think there is about an even chance that the Soviets have already [less than 1 line of source text not declassified] exoatmosphere ABM warhead yielding [less than 1 line of source text not declassified]. If they have not already done so, a Threshold Treaty would not, in our view, make such a development impossible, as we believe existing Soviet weapon technology would support it either without further testing, or with tests that would have a reasonable chance of not exceeding the threshold. (Paras. 10-12, 20, 21)
C. We believe that a Threshold Treaty would impose prohibitive restrictions, beyond those of the Partial Test Ban, only for developing weapons which might need new warheads yielding [less than 1 line of source text not declassified]. If the development of such weapons had a sufficiently high priority the Soviets might conduct tests virtually certain to violate the treaty, in the belief that the violation could not be proved against them. As few as one or two such tests a year could be of significant aid to their military programs. (Paras. 18, 19, 30-33)
D. The US Atomic Energy Detection System (AEDS) would almost certainly detect all seismic events in the USSR with a magnitude of 4.75 or greater. Perhaps with help from [less than 1 line of source text not declassified] sources, it could probably discriminate between explosions and earthquakes occurring in most parts of the USSR, but there would still be a few events a year over 4.75, especially in the Kamchatka-Kuriles area, which could not be so identified. Such events would represent possible treaty violations, but it would be extremely unlikely that intelligence could with certainty either confirm or deny that a nuclear event had in fact occurred. (Paras. 23-28)
E. If a seismic event over 4.75 was identified as an explosion, it would almost certainly be nuclear in origin. Thus this evidence, combined with what might be available from intelligence sources, would probably be sufficient, except in a few cases, to determine to the satisfaction of the US government whether or not the explosion was nuclear in origin. Evidence sufficient to convince a world forum that an explosion was nuclear could almost certainly be derived only from on-site inspection, which is not permitted by the Threshold Treaty under consideration. (Paras. 25, 29)
[Here follows the “Discussion” section, consisting of four parts entitled “Current Soviet Nuclear Weapons Technology,” “Impact of a Threshold Test Ban Treaty on Soviet Military Programs,” “US Detection Capabilities,” and “The Possibility of Soviet Covert Violation.”]
1 Source: Johnson Library, National Security File, National Intelligence Estimates, 11-66, U.S.S.R., Box 3. Top Secret; Restricted Data; Controlled Dissem. According to a note on the cover sheet, the Central Intelligence Agency, the intelligence organizations of the Departments of State and Defense, the National Security Agency, and the Atomic Energy Agency participated in the preparation of this estimate. The estimate was submitted by W.F. Raborn, Director of Central Intelligence, and concurred in by the U.S. Intelligence Board; the representative of the Assistant Director of the FBI abstained on the grounds that the subject was outside his jurisdiction.