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

486. Report of Joint U.S.–U.K. Technical Group¹

London, August 10–11, 1959

Report of Joint US–UK Technical Group to Review Technical Aspects of Nuclear Weapons Test Detection

UK Participants

• Sir William Penney
• Sir Frederick Brundrett
• Dr. R. Press
• Mr. Hulme
• Mr. Mattock
• Dr. Levin
• Dr. Maurice Hill
• Mr. Con O’Neill
• Mr. V. Macklin

US Participants

• Dr. J. R. Killian
• Dr. Hens Bethe
• Dr. Herbert Scoville
• Dr. Harold Brown
• Dr. Albert Lotter
• Dr. Carl Romney
• Dr. William Ogle
• Mr. Philip Farley

The joint group of British and American technical delegates has reviewed the technical aspects of detecting nuclear weapons tests in the light of new information which has become available since the report of the Geneva Conference of Experts and has reached the following conclusions:

(1) The group accepted for joint use by the U.S. and the U.K. the table, a copy of which is given in the appendix to this report, entitled “Estimated Capability of Control System for Detecting and Identifying Seismic Disturbances in the U.S.S.R.”

(2) The concept of the “large hole” has introduced a new kind of uncertainty which requires a re-examination of the control system agreed upon by the Geneva Conference of Experts. The “large hole” technique may provide a method of concealing underground explosions which cannot be detected by the system recommended by the Geneva Conference of Experts. This conclusion seems inevitable unless future tests contravene the theory as developed for the “large hole”.

(3) In view of the large cost and unknown feasibility of the “large hole” it is agreed that a program of engineering analysis and experimentation is desirable and necessary. The group therefore urges the prosecution and completion of high explosive tests as already planned both for the U.S. and the U.K. It further recommends that both the U.K. and the U.S. undertake engineering studies of the practicality and cost of constructing large holes and that these studies should be of such thoroughness as to give confidence in their conclusions. It is further agreed that there also should be consideration of the problems of radiation transport, of the effects of X-rays and of neurons produced by a nuclear explosion in a large cavity.
It is a conclusion of this group that planning, (and if politically unobjectionable, execution) should proceed for a series of “large hole” tests, including complete plans for a nuclear test. This planning should include the choice of sites, the instrumentation, the method of measuring yield, the time required and the cost.

(4) Consideration should also be given to the degree of decoupling achievable in “small holes”, and to alternative kinds of large cavities including the possibility of cavities including the possibility of cavities in ice and containers under water.

(5) The group notes that it is very unlikely that the engineering feasibility of decoupling by use of a “large hole” can be conclusively proven or disproven within a time shorter than a year, although high explosive tests will yield partial information within 4–6 months. Even after several years’ work the question may be unresolved. Any agreement within the next few years which bans underground tests runs some risk that decoupling schemes may be able to circumvent the experts system unless that system can be augmented in its capability by means beyond improvement of instrumentation.

(6) The group accepts the findings and recommendations of the Berkner Panel report and urges their rapid implementation. It recommends in addition that on-site inspections of selected earthquakes be undertaken as quickly as possible after these earthquakes have occurred. It is noted that such field work could appropriately be undertaken in New Zealand and California. Whenever appropriate, advantage should be taken of those underground explosions which might yield useful information for developing on-site inspection techniques.

(7) The group concluded that the only technical norms now known for the detection of explosions in large cavities would be unmanned stations of the kind discussed in the Berkner Panel report. It urges a careful technical study of the design of unmanned stations including their feasibility and associated communications and safeguarding problems. In making this comment the group emphasizes that it is not recommending at this time a political decision that unmanned stations be introduced into the Geneva discussions, since this might undermine the present proposals for manned stations.

(8) The group stresses the importance of continuing study and research for the purpose of determining new techniques for decoupling and concealment and for detection and inspection.

(9) The group recognized that there were substantial uncertainties in signal strength as a result of variations in medium, geographical and geological formations, depth of burial, etc., which in some cases could reduce the coupling below that of the Rainier event. It also recognized the difficulty of evaluating the probability that an on site inspection in the region of a concealed underground nuclear explosion would result in identification. The risk of detection may be increased by a substantial but unknown amount by the use of intelligence information in selecting the events to be inspected. The combination of all these factors makes an estimate of the capability of the system, even excluding the possible use of large holes for decoupling, very difficult below the yield level of about 20 kts. Experimentation, much of it with high explosives, could reduce some of these uncertainties. The group considers that when all the technical problems mentioned above have been fully investigated it may not be possible to offer a system which has a reliability equal to that estimated by the Geneva Conference of Experts.

(10) Although it is not possible to make a quantitative evaluation of the contribution of intelligence, it can be expected to supply assistance in determining the degree of suspiciousness of seismic events and thereby in selecting which of these events should be inspected. This assistance can be expected to be much more effective in detecting large scale unusual operations such as may be required in the construction of a “large hole”. However, intelligence by itself cannot provide a case for carrying out an inspection. Finally intelligence will introduce an element of uncertainty into any Soviet planning for a concealed nuclear test and thereby provide a deterrent to such Soviet activities.

(11) The group noted that the satellite system and ground station equipment recommended at Geneva (10 July 1959) for the detection and identification of high altitude nuclear explosions was evaluated in that Report as incapable of detecting nuclear explosions of hundred of kts. yield which are shielded against emission of X-ray radiation at a distance of more than a few tons of millions of kilometers. Such tests are a feasible means of obtaining the necessary diagnostic data on new weapons designs and might therefore, when the relative costs of conducting them are evaluated, be preferred to the large hole by a potential violator as a method of evading the control system. If one adopts the solar satellite system which is discussed but not explicitly recommended in the Report of the Technical Working Group, the capability of the systems could be improved so as to require a violator to go several times as far.

Appendix

Estimated Capability of Control System** for Detecting and Identifying Seismic Disturbances in the U.S.S.R.

		Annual Numbers of Continental Earthquakes
Equivalent Yield (KT)*		Total	Detected	Undetected	Identified	Detected but Unidentified
0.5–1		1200	750	450	5	745
1–2		675	670	5	30	640
2–5		385	385	0	85	300
Above 5		290	290	0	260	30
899

1. Source: “Large Hole;” satellite detection systems. Secret. 5 pp. Eisenhower Library, White House Office Files, Additional Records of the Office of the Special Assistant for Science and Technology, Disarmament, Nuclear Policy. Drafted on September 3.

899. * Assuming Rainier Coupling. [Footnote is in the original.]

     ** Assuming improved equipment and detection performance as suggested by the Panel on Seismic Improvement (Berkner Panel). [Footnote is in the original.]