By: Hisham Zerriffi and Arjun Makhijani, Ph.D.
May 1996

Full report [PDF 1.2MB] Summary of Findings and Recommendations As the United States participates in negotiations for a Comprehensive Test Ban treaty in Geneva, it is simultaneously putting into place a program to enhance the capabilities of its nuclear weapons design laboratories. The U.S. Department of Energy's Science Based Stockpile Stewardship (SBSS) program would build new experimental facilities to study the nuclear components of the United States nuclear arsenal, as well as a large scale computing initiative in order to more accurately model nuclear weapons. The implementation of the SBSS program is currently a part of the U.S. position to support a Comprehensive Test Ban treaty under which no nuclear explosions would be permitted (the "zero yield" CTB). Officially, the purpose of the SBSS program is to maintain the safety and reliability of the nuclear arsenal in the absence of nuclear testing. The Department of Energy (DOE) has argued that, as the nuclear arsenal ages, it will be an increasingly complex task to maintain the level of safety and reliability necessary without nuclear testing. The SBSS program is supposed to aid in this endeavor by providing information on the basic physical processes of nuclear weapons in order to create more accurate computer models, in essence to be able to conduct "virtual tests." The purpose of this report is to examine DOE's statements regarding the safety and reliability of the nuclear arsenal and to assess the usefulness of the SBSS program in addressing the relevant issues. Our analysis of historical data regarding problems with nuclear warheads leads us to conclude that the SBSS program would provide little aid in maintaining the safety of the existing arsenal. Indeed, DOE's own data show that there have been no aging-related nuclear safety problems in warheads. While the SBSS program's claims in regard to improving safety of the arsenal appear dubious at best, it has a clear relationship to increasing U.S. capability to design new warheads and to design major modifications to existing ones. The new SBSS facilities are of the types used previously as part of the weapons design program. One of the main goals of the program is to retain and attract new weapons designers. Furthermore, various official documents indicate that the ability to maintain weapons design capabilities is a priority of the DOE. Another purpose appears to be to maintain the reliability of the nuclear arsenal at extremely high levels. Such high levels of reliability may be necessary only if the United States pursues a strategy of first strike against opponents with large nuclear arsenals rather than retaliatory nuclear deterrence. However, the data that we have are too limited to enable us to arrive at a definitive conclusion in this regard. The SBSS program is coupled with other problematic provisions in the U.S. position on the CTB. (1) Specifically, the U.S. government wants to: maintain the Nevada Test Site in a state of permanent readiness to resume full scale testing; have a provision that would allow withdrawal from the CTB for reasons of "supreme national interest." The design capabilities inherent in existing and new SBSS facilities will provide the opportunity for the DOE to bring new weapons or modifications to existing weapons to a stage of near completion, in the same manner that a complex machine such as the Boeing 777 was largely designed using computers and wind tunnels. The SBSS program is likely to create pork-barrel driven pressures to withdraw from the CTB in times of crisis. The enormous financial advantages that the U.S. enjoys over Russia and China in the matter of military expenditures, despite recent reductions in the U.S. military budget, could contribute to reluctance on the part of other powers to engage in nuclear arms reductions. This analysis leads us to the conclusion that a large SBSS program which includes expensive new experimental facilities with weapons design capabilities, could lead to dangerous international instabilities. It could have profound negative repercussions on the functioning of both the Nuclear Non-Proliferation Treaty (NPT) and the upcoming Comprehensive Test Ban treaty (CTB). Principal Findings The Department of Energy's analysis of the need for an SBSS program mixes up safety and reliability issues in a misleading way. These issues are technically distinct and have vastly different political, military, and environmental implications. So long as there are intact warheads, nuclear safety is an issue of the greatest concern, since the human and environmental consequences of accidental detonations could be devastating. Reliability of warhead performance is a technical and political issue that is linked to military strategy. DOE has not justified the need for SBSS facilities as they relate to safety separately from reliability issues, and it has not related levels and types of reliability required to military strategy. DOE data show that the nuclear detonators in nuclear warheads (called "primaries") have never had safety problems linked to aging. The data clearly indicate that SBSS facilities are not needed for nuclear safety. DOE statements that imply that aging-related nuclear safety issues can be solved using the SBSS program are not well-founded either in data or in analysis. Similar statements claiming the need for new facilities are even less justified. The SBSS program will give the U.S. powerful capabilities for designing new warheads (as mandated by present nuclear weapons policy). While these capabilities are unlikely to allow DOE to bring radical new warhead designs into production for deployment by the Pentagon, they would allow most design work to be completed, and the rest to be rapidly concluded should the U.S. withdraw from the CTB. These principal findings are based on our detailed analysis of DOE data on safety and reliability problem types that is presented in the report. The main technical points in our analysis are as follows: The majority of kinds of safety and reliability problems have arisen from design or production of the warhead, rather than aging. As a result, the majority of problem types are found within the first few years of a warhead's production. Only 12 percent of safety problem types have involved aging, and only one-fourth of these were found in warheads still in the current stockpile. These have been addressed. All safety problems with primaries -- the most crucial component for nuclear weapons safety -- have been the result of the design of the warhead, rather than aging. The principal means of finding defects has been the Stockpile Evaluation Program, which does not involve the experimental facilities that are part of the DOE's SBSS program. Hydrodynamic testing, which involves many existing and proposed SBSS facilities, has a role in helping determine "one-point safety," which is a basic safeguard against accidental detonation. However, the DOE has already certified existing warheads as safe in this regard. Therefore, it would appear that even existing hydrodynamic facilities may not have any further role on one-point safety, especially as historical data do not indicate any aging-related nuclear safety problems. In view of its own declaration that the existing stockpile is safe and of its own data regarding the lack of aging-related nuclear safety defects, the DOE has not made a case that new hydrodynamic testing facilities will make material contributions to safety. High energy density facilities, such as the $1 billion National Ignition Facility, would have no relevance to maintaining the nuclear safety of existing weapons in the arsenal. While aging has a greater affect on reliability than safety, reliability problem types are primarily with non-nuclear components and rarely have a severe effect on a warhead. While there may be deleterious effects on reliability if the stockpile is held for very long periods, these problems could be addressed without the SBSS program by re-manufacturing the defective parts. What is important is the final re-manufactured product, not the particular industrial process used for warhead maintenance. This process should be workable so long as there is no attempt to "improve" warhead design as part of re-manufacturing. The weapons effects component of the SBSS program is relevant to warhead effectiveness, not warhead safety. The manufacture of new nuclear components that are significantly different from the tested original could result in less reliable and/or less safe warheads. Discussion of Principal Findings: Maintaining the safety of nuclear weapons should be one of the top priorities of the DOE's nuclear weapons complex as long as intact nuclear weapons remain in the arsenal. Accidental nuclear detonation or plutonium dispersal could have huge health and environmental consequences. However, the DOE has not demonstrated the need for the SBSS program to maintain nuclear safety. Nuclear weapons are currently safe, according to the DOE. Safety problems with primaries have never been linked to aging. Furthermore, 76 percent of the safety-related problem types in primaries were found in warheads produced around the time of the 1958-1961 U.S.-Soviet nuclear testing moratorium, a time of rushed design work as the United States scrambled to get designs into production. Nuclear weapons are currently reliable, according to the DOE. The majority of reliability problem types affect non-nuclear components and the majority of reliability problem types have a minimal effect on the warhead. In light of these facts, the DOE has failed to state how the SBSS program would maintain the reliability of an arsenal for a policy of retaliatory nuclear deterrence -- that is, a policy of nuclear retaliation in response to first use of nuclear weapons by an adversary. However, if the purpose of the arsenal is a first strike, a higher degree of reliability may be necessary, because achieving precision and rated yield could be technical factors affecting the "success" of a first strike aimed at destroying an adversary's nuclear missiles. Maintaining the ability to design and produce new nuclear weapons, or even making militarily significant modifications to existing warheads, could have serious implications for the Comprehensive Test Ban Treaty and nuclear non-proliferation. The design capabilities of the SBSS program coupled with maintaining the Nevada Test Site in a state of readiness to resume testing could lead to serious new instabilities with unforeseeable consequences. For instance, it could lead to the disintegration of a Comprehensive Test Ban. It also violates the spirit of Article VI of the NPT under which the nuclear weapons states are pledged to pursue nuclear disarmament negotiations in "good faith." The DOE has not considered alternative approaches to maintaining the safety of the nuclear arsenal after a CTB in its Draft Programmatic Environmental Impact Statement for Stockpile Stewardship and Management. There are viable options which would not require new experimental facilities or the maintenance of a large cadre of weapons designers. Recommendations DOE should demonstrate, in light of its own historical data on nuclear safety, why new experimental and computational capabilities are relevant to the safety of the existing U.S. nuclear arsenal. The U.S. should adopt a policy of dismantling warheads whose primaries are deemed to be unsafe, instead of a policy that would make changes to the "physics package," which is the nuclear portion of warheads. This appears more prudent from the point of view of safety. It would also be in keeping with the spirit of the commitments of the nuclear weapons states under Article VI of the Non-Proliferation Treaty. The U.S. government, including the DOE and DoD, should address specifically how the SBSS program is relevant to a strategy of retaliatory deterrence as distinct from a first use and first strike nuclear strategy. The option of first use of nuclear weapons has historically been part of U.S. nuclear strategy. (2) The U.S. government should clearly and unambiguously renounce nuclear weapons design and development, and invite international verification of this policy. It should also use the leverage created by the unilateral adoption of such a policy to pressure the other nuclear powers to follow suit. Before embarking on the SBSS program, the DOE and DoD should examine carefully the ways in which it could create dangerous new international instabilities, including in U.S. relations with Russia, China, and the undeclared nuclear weapons states, Israel, India, and Pakistan. Before embarking on the SBSS program, the DOE should examine carefully the non-proliferation consequences of the SBSS program. This examination should include the possible relation of the SBSS program to the potential for the U.S. or other countries breaking out of the CTB, and the potential for a breakdown in the CTB altogether. In order to further the possibility of achieving a CTB in 1996 in a manner that could lead to greater security and greater confidence that the nuclear powers intend to create a path to nuclear disarmament and eventually abide by the spirit of Article VI of the NPT, the U.S. should: Permanently close down the Nevada Test Site and focus on clean-up and conversion. Use the leverage gained by shutting down the Nevada Test Site to pressure other nuclear powers to permanently shut down their nuclear test sites. Cancel all underground sub-critical experiments. Halt construction of new facilities under the SBSS program and on that basis urge the other nuclear weapons states to halt construction of similar facilities. Participate with other countries to create a treaty against the first use of nuclear weapons in any conflict and against any threat of use of nuclear weapons against non-nuclear weapons states. Endnotes 1 As noted in the preface, the other nuclear weapons states are pursuing similar policies. 2 Ellsberg 1986.