Foreign Relations of the United States, 1961–1963, Volume VII, Arms Control and Disarmament

128. Memorandum From the Deputy Secretary of Defense (Gilpatric) to President Kennedy⁰

Washington, undated.

SUBJECT

• Resumption of Atmospheric Nuclear Testing

It is the Department of Defense position that immediate military requirements exist which provide a clear basis for resumption of atmospheric nuclear testing at this time. Of particular importance are weapon effects data which are critically needed to assure that both offensive and defensive weapon systems are and continue to be effective.

Weapon development tests in the atmosphere would provide increased diversity for existing and future weapon systems in the face of changing and sometimes unpredictable threats. The particular weapon development experiments in the proposed series are unlikely by themselves to produce radical changes in our relative strength. However, the developments resulting from several test series will certainly produce an important difference in military capability. Furthermore, prudence demands we be alert to the possibilities of major breakthroughs which can only be explored with a vigorous basic weapon development program including atmospheric testing.

Operational tests of existing systems, though less critical than effects and development tests from a purely technical viewpoint, are highly recommended if the decision is made to test in the atmosphere. Such full scale tests of systems would go a long way toward strengthening the confidence of the Congress and the public in the forces on which we place so much reliance in our defense and war plans. The reliability of our deterrent forces should be credible to ourselves as well as to our potential enemies.

(1) Weapons Effects

Preparations are being made for four specific weapons effects tests for early execution: two at high altitude, one on the surface in Nevada, and one underwater. These tests by no means satisfy the total needs, but will, even if no others are subsequently carried out, make a qualitative change in our ability to understand and predict such effects, and to make changes in our operational forces to take account of them. It should be noted that though these tests will probably raise as many questions as they answer, our military capability and flexibility will be improved greatly even by the posing of the unanswered questions. Preparations and planning are also underway for overseas tests which could build on the results of a first series and be ready about twelve months later, as well as additional small-scale tests in Nevada.

The first series of high altitude tests will provide critical data for both offensive and defensive systems. In the absence of an active enemy defense, existing systems are adequate to insure unacceptable damage to any enemy. However there is clear evidence that the U.S.S.R. is working vigorously on AICBM development. To counter this possibility we must minimize the vulnerability of our delivery systems and provide measures to penetrate their defenses. The high altitude tests will provide critical data necessary to evaluate nuclear “blackout” and associated electromagnetic effects. They will also provide some additional information on the vulnerability to enemy attack of U.S. nuclear warheads and reentry vehicles at two altitudes of interest. With these data we can provide inputs for the design of systems to aid in the penetration of potential enemy defenses.

The same kinds of information are essential to an evaluation of defensive problems. The attacker may attempt to “blackout” our AICBM systems and in any case we may have “blackout” problems from our own defensive explosions. To evaluate and optimize the kill effectiveness of our possible AICBM systems we must, for lack of precise knowledge of the enemy warhead, obtain and use basic information on the vulnerability of U.S. systems.

High altitude nuclear bursts produce widespread and long-lived disruption of certain radio communication systems. Considerations of command and control demand that we understand the magnitudes of the effects and adjust to the possible loss of communication channels on which we now place heavy reliance.

The low yield surface effects test planned for Nevada will provide important information on the vulnerability of hardened missile sites to enemy attack. This experiment will tell us whether the electromagnetic signal induced by the explosion can damage command and communication links in varied ICBM sites, and give us an experimental point in calculating whether our hardened sites will survive nearby multi-megaton explosion. This test will also provide important data on blast and shock relevant to estimating effects on hardened structures. Certain warhead vulnerability experiments, pertinent to AICBM problems but best performed on the surface, will also be incorporated in this test.

The underwater test is both an effects test and a systems proof test. The [less than 1 line of source text not declassified] ASROC warhead will be detonated at a depth of 750 feet. An example of the critical effects information that defies simulation or calculation is the response of the complex sonar gear to such an explosion. It is probable that in a combat situation more than one ASROC would be fired and it is important to determine how accurately if at all track can be maintained on hostile targets. In addition information will be obtained on the fleet operational problems in a nuclear environment of blast and radioactivity.

(2) Weapons Development.

The planned AEC weapon development shots fall in three general categories:

(a) Tests of devices designed during the moratorium and now entering stockpile. Each specific device of this category represents the maximum acceptable extrapolation from tested devices and because of the critical nature of the systems they are part of they should be tested. Further these tests will provide normalization of the design calculations for the more advanced devices also proposed for testing. If the advanced designs do not perform as expected this normalization data will be essential for understanding the reasons.

(b) Tests of advanced designs providing increased yield-to-weight and/or greater diversity. These designs represent in general extrapolations beyond those acceptable for certification without testing. Present devices delivered on target would be adequate. However, in the face of an active enemy defense these advanced designs will make major contributions to the penetration problem. Our possession of multiple warheads [less than 1 line of source text not declassified] would make it much more difficult for the Soviets to produce systems able to intercept the variety of penetrating attacks we could then mount.

One basic point about yield and penetrability needs to be understood. If a single ballistic missile warhead has a yield of less than a megaton the enemy has a real advantage in AICBM design because to be effective the incoming device must be detonated at low altitude. This gives the enemy AICBM system maximum time and use of the slowing-down effects of the atmosphere to aid in discrimination and intercept in the face of whatever penetration tactics we use. Increasing the yield at a fixed weight will make detonation at a higher altitude possible for existing systems such as Minuteman and Titan II, thus increasing the enemy AICBM problem.

Finally, some counter-force capability seems desirable even with a second strike strategy. Large yields then become essential and must be compatible with existing weight capabilities of our systems.

(c) Developmental tests exploring major new design concepts. These tests represent first steps toward major improvements in weapon performance. Being first steps they will undoubtedly require further testing at future times to exploit the knowledge gained in this series. If we wish to maintain a vigorous weapon development program over the next decade these experiments are essential. The two groups of tests previously discussed are addressed to immediate problems, whereas this group look to the future. Weapon development, like any exploratory area, takes time and usually contains many surprises. A few of the proposed experiments are thus of an exploratory nature rather than oriented toward immediate or easily foreseen requirements.

(3) System Tests.

It appears possible and highly desirable to incorporate in the planned test program proof tests of the complete Polaris and Atlas systems. These tests will not jeopardize either the weapon effects or weapon development program since they can be carried out almost completely independently of the others. It is recommended that these system tests be included in the planned series; they have a total [less than 1 line of source text not declassified].

It is a matter of judgment how much testing, and of what kinds, is necessary for confidence in system performance. Plans are underway for exhaustive user-type missile tests. Missile tests have been carried out measuring the environment to which the nuclear warhead is subjected, and verifying that the separately tested nuclear warheads will perform after missile flight. Nevertheless, the Joint Chiefs of Staff believe that the proposed systems tests are essential to building confidence in our major existing deterrent forces and to confirm the operability of complete weapon systems. They are submitting a paper outlining in more detail the basis of their requirements.¹ Mr. McNamara and I concur in the desirability of complete system tests of selected systems.

The Joint Chiefs of Staff at the same time support the critical need for both the weapon effects and weapon development programs as outlined above.

(4) Effect of Not Resuming Atmospheric Testing on Future Weapons Development and Weapons Effects Capability.

It is in the nature of weapon effects and weapon development research that one must rely heavily on theoretical studies. But theory without experimental confirmation and guidance soon becomes sterile. We are now at a point in important areas of such research where full scale experiments are absolutely essential to any further progress.

Even ignoring the immediate military requirements that motivate most of the proposed atmospheric tests, this serious loss of momentum and capability with time should weigh heavily in a decision to resume atmospheric testing. A vigorous and reasonably unfettered weapon program is the best insurance against unpleasant and unpredictable surprises of the future. If the decision is made not to conduct atmospheric tests during 1962, and if the Soviets carry out another extensive test series in 1963 or 1964, we would at that time be in a substantially worse position than we are now with respect to both state and momentum of development compared to the Soviets.

0. Source: Kennedy Library, National Security Files, Subjects Series, Nuclear Weapons Tests, 497th NSC Meeting. Secret; Restricted Data.
1. Memorandum from the Joint Chiefs of Staff to the President dated February 16. (Ibid.) See the Supplement. Also on February 16, Seaborg submitted to the White House an unaddressed memorandum reviewing objectives and operational plans for the proposed test series, and detailing changes in the program since his November 29 letter (Document 102). (Kennedy Library, National Security Files, Subjects Series, Nuclear Weapons Tests, 497th NSC Meeting)