Foreign Relations of the United States, 1958–1960, Volume III, National Security Policy; Arms Control and Disarmament, Microfiche Supplement

565. Memorandum for the Record by Kistiakowsky¹

June 6, 1960

SUBJECT

• Geneva Nuclear Test Negotiations

As I shall be accompanying the President on his Far Eastern trip, I have prepared the following comments on several technical aspects of the Nuclear Weapons Test Cessation problem as guidance to Mr. Keeny of my office in the event that he attends a Principals meeting as an observer during my absence.

I. The Underground Detection Problem.

The Soviet delegates in Geneva have refused to agree with our plans to include decoupled explosions as well as small (below about 1 KT) tamped nuclear explosions in the seismic research program, claiming that these do not contribute to the improvement of the system and/or are a subterfuge aimed at the resumption of development of small tactical weapons.

While I strongly feel that we should continue with a seismic improvement program, if necessary on a unilateral basis, I have misgivings about proceeding on a unilateral basis with the present Vela program, particularly with safeguards based on the “black box” approach. My reasons for this misgiving are that the nuclear tests in the Vela program discussed in paragraphs (a) to (c) below are technically not all indispensable and could be modified to meet some of the Soviet criticisms without damaging seriously our program. Therefore, if we proceed on this program with “black box” safeguards, we may invite world-wide criticism that we are using a subterfuge in order to resume weapons tests. This possibility should certainly be taken into account before deciding to proceed with the present Vela program of nuclear explosions.

In considering this problem, it should be noted that the Ad Hoc Panel which reviewed the Vela program, presented by us at Geneva, was not instructed to reduce the number of nuclear explosions to the technically indispensable minimum and therefore accepted AEC–DOD plans for twelve nuclear explosions as contributing technically to the program. Several component issues are involved here, which I will consider separately below.

(a) Tamped explosions. The program to study the effect of environment and yield on tamped nuclear explosions includes shots of 1 KT and ¼ KT yield. I discussed the matter with three members of the above mentioned Ad Hoc Panel; they feel that these shots could be done chemically or even eliminated without seriously impairing the program, but emphasize that some other members of the panel might well take a different view. It is an issue on which no technical unanimity is likely to be achieved.

A special problem is presented by the earliest scheduled shot, “Lollipop,” (5 KT) to be fired in September. It has come to my attention that the Department of Defense is constructing extensive underground installations in close proximity to the locus of this explosion for “weapons-effects tests,” (i.e. to determine the effects on different types of construction for hardened missile sites). Whether this use of the seismic improvement program would be accepted by world opinion, I do not care to estimate, but am sure that the existence or intent of the installations could not be kept secret, especially if UN or USSR observers are present there.

(b) Decoupled explosions. In addition to one large (10 to 20 KT) partially decoupled explosion, the program includes 0.1 KT, 0.5 KT and 2.0 KT decoupled or partially decoupled explosions as well as a [illegible in the original] KT tamped calibration explosion. The principal reasons for the small yields chosen are that smaller underground holes can be used, reducing construction time and cost. However, the explosions are so small that they might yield only incomplete technical information relating to possible methods of detecting and identifying decoupled explosions. It should be emphasized that the Geneva system, even after substantial improvements and some expansion, cannot be expected to detect such small decoupled explosions. Therefore, unfriendly minds could denounce them as a program for developing techniques for evasion of monitoring. The panel members with whom I discussed the issue feel that, for instance, a single, fully decoupled 5 KT shot would be more instructive as far as possible improvements are concerned than the present program, but they emphasize that this will entail a significant delay and that such a modified program will probably be objected to by some members of the Panel. It is probable that the Panel would be unanimous in insisting from the technical point of view that, as a minimum, one or two decoupled shots of perhaps 5 KT be included in the program, as otherwise no information on means of detection of decoupled explosions will be forthcoming because chemical explosives cannot be used for this purpose.

(c) Safeguards for nuclear explosions present a serious problem because of the flat rejection of the “black box” plan by the Soviets. If our program is reoriented to include only explosions of a size of a few KT and larger, it should be possible, technically and I believe legally, to resort, if a compromise is sought, to one of the following schemes: (i) to use UK weapons which were developed prior to the US–UK bilateral agreement. These could be inspected by US and USSR on UK soil, then sealed into “black boxes” and imported into USA. (ii) to use devices based on the old US “gun type” design, which could be inspected by the USSR and UK after declassification. As this design requires a large amount of U–235 and is extremely inefficient, its declassification would be of only very limited value to any nation attempting to join “the club.” The cost to us of these devices will be small compared to the total costs of the Vela program, although larger than in the case of implosion weapons. Whether the political problems involved in either of these proposals can be resolved, I do not know.

(d) The estimate of the cost of the monitoring system in the USSR was mentioned at the Joint Committee hearing and at the Principals meeting as $1–5 billion. I have not been able to discover any foundation for this figure. The relevant report of United Electrodynamics to AFTAC gives only the figure of $75 to $128 million for the installation of arrays of seismometers at 22 stations in the USSR. It does not include other costs, such as logistic support system, other equipment, housing and communications to the World Control Center at Vienna. I understand that United Electrodynamics has recently unofficially estimated the total cost for USSR at $750 million (which included allowance for nuclear testing of system). This figure is much in doubt and is much higher than the corresponding figure given in the estimate prepared by AFTAC.

II. The High Altitude Detection Problem.

As you may recall, the experts in Geneva agreed on the feasibility of monitoring high altitude weapons tests through a fantastically complex system, including ground-based instruments, low-altitude satellites, high-altitude satellites and solar satellites. The DOD briefing given to the Principals presented an especially pessimistic (and not wholly correct) assessment of some features of this plan. I understand that certain of the pessimistic conclusions on the time of availability of X-ray detection are now being modified as a result of a review by the OSD Ad Hoc Committee on High Altitude Detection. More significantly, however it now appears that the cost of a satellite high altitude detection system may be greatly reduced as a result of work by Lockheed Corporation. This idea makes use of the fact that fission fragments from explosions relatively close to the earth would be mechanically trapped in the upper atmosphere and would radiate gamma rays for an appreciable time. This would permit the substitution of four low-altitude (500 km) satellites for a larger number of high-altitude satellites. Satellites of this type with somewhat simplified X-ray equipment for deep space detection could probably be available in two or three years. When more sophisticated X-ray equipment becomes available, this system would have essentially the same capabilities as the high-altitude satellite system described by the DOD. The cost for this type system would be in the $100 million class rather than in the $1 billion class. A definitive analysis of this plan should be available by mid-June, but it looks most promising. I might add that low-altitude satellites were a part of the Geneva plan and the new scheme merely makes the rest of the plan unnecessary, unless one wants to go to the so-called “solar satellites” to monitor nuclear explosions behind the sun and to force a violator to more complicated shielding devices. My guess is that one could have a moratorium on such tests in deep space for quite a few years without fear of evasion, because of the great costs and difficulties of carrying out such weapons tests.

1. Source: Comments on technical aspects of nuclear weapons tests cessation. Secret; Eyes Only. 4 pp. Eisenhower Library, White House Office Files, Additional Records of the Special Assistant for Science and Technology.