C.T.B.T. Truth or Consequences #7: Realistic Explosive Testing is Required to ‘Remanufacture’ Existing Nuclear Weapons

(Washington, D.C.): One of the most pernicious misrepresentations being served up in
recent
days by the proponents of the Comprehensive Test Ban Treaty (CTBT) is the claim that the U.S.
deterrent stockpile can be maintained for the indefinite future without further underground tests.
Since they explicitly rule out modernization of the nuclear arsenal, however, the only
way a
stockpile comprised of weapons having the highest average age in history could possibly be
preserved in a safe and reliable condition would be if existing weapons types were to be
substantially (if not virtually completely) remanufactured.

While advocates of the zero-yield, permanent CTBT deny it, neither historical experience
and
common sense support the proposition that U.S. nuclear weapons — comprised as they are of as
many as 6,000 exactingly manufactured parts, made of exotic and often dangerous materials and
constantly exposed for years to high levels of radiation — will not undergo substantial changes
over time. In fact as a result of such factors, former Assistant Secretary of Energy Victor Reis
declared in congressional testimony in October 1997 that: “Just about all the parts [of
these
obsolescing devices] are going to have to be remade.”

Why ‘Remaking’ of the Arsenal Cannot be Effected Without
Testing

There a numerous, serious problems with undertaking such a program in the absence of
nuclear
testing. First, the production lines for building the stockpile’s existing bombs
and warheads
were disassembled long ago. Reconstituting and recertifying them would take quite some time,
would be very costly and probably won’t be possible to effect with confidence absent realistic,
explosive nuclear testing.

Second, it will not be possible to replicate some of the ingredients in
weapons designed two
decades or more ago; key components are technologically obsolete and no one would
recommend using them when smaller, lighter, cheaper, more reliable and carcinogenic materials
are now the state-of-the-art. In addition, federal safety and health guidelines prohibit the use of
some of the materials utilized in the original designs.

Third, virtually everybody who was involved in designing and proving the original
designs
has left the industrial and laboratory complex, taking with them irreplaceable corporate
memory that may spell the difference between success and failure in reproducing their
handiwork.

An Authoritative Historical Review

These points were underscored in an authoritative report to Congress issued by the Lawrence
Livermore National Laboratory in 1987. Among its relevant highlights are the following
(emphasis added throughout):

• “It has frequently been stated that non-nuclear and very-low yield (i.e., less than 1 kiloton)
  testing and computer simulation would be adequate for maintaining a viable nuclear deterrent.
  A recent variant of this argument asserts that while such testing and computer simulation may
  be insufficient for the development of new warheads, they would be adequate for
  indefinite
  maintenance of a stockpile of existing weapons. We believe that neither of these
  assertions
  can be substantiated.

“The major problem is that a nuclear explosive includes such a wide range of
processes and scales that it is impossible to include all the relevant physics and
engineering in sufficient detail to provide an accurate representation of the real
world.”

• “A final proof test at the specified low-temperature extreme of the W80 (Air-Launched
  Cruise
  Missile) was done as the weapon was ready for deployment. The test results were a
  complete
  surprise. The primary gave only a small fraction of its expected yield, insufficient to
  ignite
  the secondary.

“Our experience with the W80 illustrates the inadequacy of non-nuclear and
low-yield testing and the need for full-scale nuclear tests to judge the effects of small
changes. Even though it has been argued that such a “thorough” test should have
occurred earlier, the critical point is that computer simulation, non-nuclear testing, and
less-than-full-scale nuclear testing are not always sufficient to assess the effects of
deterioration, changes in packaging, or environmental conditions on weapon
performance.”

• “Testing of newly produced stockpiled systems has shown a continuing need for
  nuclear
  tests. Even an “identical” rebuild should be checked in a nuclear test if we are to
  have
  confidence that all the inevitable, small and subtle differences from one production run
  to the other have not affected the nuclear performance. The current stockpile is
  extremely
  reliable, but only because continued nuclear testing at adequate yields has enabled us to
  properly assess and correct problems as they occur.”

• “Although tests of a complex system are expensive and time-consuming, one is hard-put to
  find an example anywhere in U.S. industry where a major production line was reopened and
  requalified without tests. Exact replication, especially of older systems, is
  impossible.
  Material batches are never quite the same, some materials become unavailable, and equivalent
  materials are never exactly equivalent. Different people–not those who did the initial work–do
  the remanufacturing.

“Documentation has never been sufficiently exact to ensure replication. A perfect
specification has never yet been written. We have never known enough about every
detail to specify everything that may be important.

“Tests, even with the limitations of small numbers and possibly equivocal
interpretation of results, are the final arbiters of the tradeoffs and judgments that
have been made. We are concerned that, if responsible engineers and scientists
were to refuse to certify a remanufactured weapon, pressures could produce
individuals who would. The Challenger accident resulted from such a situation
and highlights an all-too-common tendency of human nature to override
judgement in favor of expediency.”

• “Remanufacture of a nuclear warhead is often asserted to be a straightforward exercise in
  engineering and material science, and simply involves following well-established
  specifications to make identical copies. In the real world, however, there are many
  examples where weapon parts cannot be duplicated because of outmoded technologies,
  health hazards, unprofitable operations, out-of-business vendors, irreproducible
  materials, lack of documentation, and myriad other reasons….Not only must
  remanufacturing attempt to replicate the construction of the original weapon, it must also
  duplicate the performance of the original weapon.”

• “It is important to emphasize that in weapon remanufacture we are dealing with a
  practical problem. Idealized proposals and statements that we ‘should be able to
  remanufacture without testing because expertise is not essential’ are a prescription for
  failure.”

The Bottom Line

Senators concerned about the Nation’s ability to perform the needed modifications essential
to
any effort to “remanufacture” stockpiled weapon types should bear in mind a comment by one of
the prominent scientists usually cited by CTBT proponents: Dr. Richard
Garwin. In testimony
before the Senate Foreign Relations Committee last week, Dr. Garwin declared: “I
oppose
modifying our nuclear weapons under the moratorium or under the CTBT.”

Given historical experience and the scientific insights gleaned from it, no one who is serious
about maintaining the U.S. deterrent for the indefinite future would argue that the existing
inventory can be perpetuated without nuclear testing. Remanufactured weapons will have to be
realistically tested, at least at low-yield levels, if we — and those we hope to deter — are to have
confidence in their effectiveness.