For the U.S. nuclear weapons program, the primary concern related to the Comprehensive Nuclear Test Ban Treaty (CTBT) is whether confidence in the safety, security,1 and reliability of the weapons can be maintained for the foreseeable future (see Box 1-1) without nuclear-explosion testing. In this chapter we address this issue, emphasizing what has changed since the 2002 Report (NRC, 2002). We begin with an overview, including the relevant findings of the 2002 Report, and then describe key changes since that report. This is followed by a summary of the current status and a description of what must be done in the future to maintain a safe, secure, and reliable stockpile without nuclear-explosion testing.
BOX 1-1 The Committee’s Judgments about the Future
A treaty of indefinite duration must be evaluated taking into account possible changes over time. Throughout this report, we frequently use the phrase “foreseeable future” in making judgments about the future. By this term, we mean that we have been unable to identify future developments that would alter our judgments and recommendations (In some cases, we caveat a specific judgment by noting that it applies only if our recommendations are followed). At the same time, we are not claiming omniscience about an inherently uncertain future. Because it is possible that conditions we have not foreseen will arise, it is important that the United States establish and maintain appropriate safeguards and, if necessary, be prepared to exercise the withdrawal provisions under the CTBT’s supreme national interest article.2
In Chapter 1, our use of the phrase “foreseeable future” signifies our assessment that the United States can maintain its existing nuclear weapons with certain safeguards over the long term. This is based on our intimate knowledge of the U.S. nuclear weapons program past, present, and likely alternative futures.
On the other hand, in Chapter 4 we present technical constraints imposed by monitoring and verification on actions that other countries might or might not take. Depending on the country, history shows that this could change over the near term. Thus, we use phrases such as “it is likely that” to leave a healthy margin for such uncertainty (We know what we know, but not necessarily what we do not know). Even so, we conclude that with adequate U.S. safeguards, the actions taken by other countries, although not as predictable as our own, will not likely change U.S. nuclear weapons policy or impose requirements for new weapons “for the foreseeable future.”
1 The nuclear weapons community uses the term surety as an umbrella term covering safety, security, and use control of nuclear weapons. Because the technical details of use control exceed the security classification of this report, the committee does not address use-control issues in the report. To make this clear, the committee uses the term safety and security rather than surety throughout the text.
2 CTBT requires a 6-month delay after invoking the supreme national interest clause before a State could conduct a nuclear-explosion test. However, the time needed to prepare a test would be greater than 6 months.
At the turn of the century, the United States had been observing a self-imposed nuclear-explosion test moratorium for more than seven years and was in the early stages of learning how to maintain its nuclear arsenal in the absence of nuclear-explosion testing. Concerns about the feasibility of this task were an important factor in the 1999 Senate decision not to give its advice and consent to ratification of the CTBT.
The 2002 Report provided important background and tutorial material on many topics relevant to nuclear weapons. This included an historical perspective on nuclear-explosion testing; a description of the origin of the Stockpile Stewardship Program (SSP); a discussion of the process by which a warhead enters the stockpile and is then maintained; and the factors influencing the safety, security, and reliability of the weapons.
The focus of these sections was on the ability to maintain the existing stockpile in the absence of nuclear-explosion testing. The implied premise was that the military requirements were frozen and that each system would be maintained in a form as close to the original military specifications as possible. Under these assumptions the assessment in the 2002 Report was that the safety and reliability of the U.S. stockpile could be maintained via careful adherence to past practices and that six measures were most important to accomplishing that purpose:
We judge that the United States has the technical capabilities to maintain confidence in the safety and reliability of its existing nuclear-weapon stockpile under the CTBT, provided that adequate resources are made available to the Department of Energy’s (DOE) nuclear-weapon complex and are properly focused on this task. The measures that are most important to maintaining and bolstering stockpile confidence are:
• maintaining and bolstering a highly motivated and competent workforce in the nuclear-weapon laboratories and production complex
• intensifying stockpile surveillance,
• enhancing manufacturing/remanufacturing capabilities,
• increasing the performance margins of nuclear-weapon primaries,
• sustaining the capacity for development and manufacture of the non-nuclear and nuclear components of nuclear weapons, and
• practicing “change discipline” in the maintenance and remanufacture of the nuclear subsystems. (NRC, 2002, pp. 1, 9.)
The 2002 Report offered many cautionary notes about the risks associated with any change from the initial configuration of the warhead, especially in the nuclear explosive package. The report was quite positive on the ability to maintain the stockpile without nuclear-explosion testing as long as these risks were respected and the six measures were followed.
The U.S. Stockpile Stewardship Program
At the time of the 2002 Report, there was uncertainty about the nascent SSP and maintaining the stockpile in the absence of nuclear-explosion testing. The intervening 10 years have seen major successes in the discovery and resolution of significant stockpile issues, as well as notable problems in maintaining the physical and human infrastructure needed for the SSP. The successes indicate that it is possible to maintain a safe, secure, and reliable stockpile
into the foreseeable future without nuclear-explosion testing. The problems indicate that future success would be in jeopardy without a continuing commitment to stewardship and the nuclear weapons complex. The issues concerning the sustainability of the enterprise are discussed in Chapter 3 of this report.
A number of changes have occurred since the 2002 Report was written, some technical and others political with technical implications. Here we discuss some of the most important ones:
• There have been significant advances in technical knowledge and capability since the 2002 Report. These include:
studies based on work done at Los Alamos National Laboratory (LANL) and Lawrence Livermore National Laboratory (LLNL) on plutonium (JASON, 2007), which set a lower limit on pit lifetimes of 85-100 years and which include the resolution of difficult materials issues;
the development of peta-scale computational capability and its application to both design and stockpile problems;
completion of the National Ignition Facility and initial experiments involving a megajoule of laser energy; o completion and operation of other SSP-related major research facilities such as the Dual-Axis Radiographic Hydrodynamic Test Facility (DARHT) and the Microsystems and Engineering Sciences Application Facility (MESA); and
successful solutions to a number of weapons issues found during surveillance and design work.
These technical competencies are reflected in the annual letters by the DOE nuclear weapons laboratory directors and the Commander of U.S. Strategic Command (STRATCOM), who conclude that a nuclear-explosion test is not presently necessary to maintain the continued safety, security, and reliability of the stockpile.3
• Production of certified pits has been demonstrated at Los Alamos, in particular for the W88. This is an important milestone and provides the technical basis for meeting future manufacturing and remanufacturing requirements.
• The events of September 11, 2001, occurred too late to have any impact on the 2002 Report. There have been several subsequent consequences for the nuclear weapons enterprise, some direct and others indirect. First, the skills of people knowledgeable about nuclear weapons have been sought for insight into ways to prevent or mitigate a terrorist nuclear threat. Second, there is increased interest in incorporating additional safety and security features in stockpile weapons, including those requiring changes to the nuclear explosive packages. Further, there have been substantial upgrades of the security posture at sites within the nuclear weapons complex to guard against potential terrorist threats (with concomitant budgetary and operational impacts on the conduct of work).
• The W87 underwent a successful life-extension program (LEP) to improve its long-term integrity (NNSA, 2004). The nuclear explosive package was modified to improve weapon reliability.
3 The first letters were written in 1996. The Annual Assessment process was established as a result of President Clinton’s statements on August 11, 1995, when he announced the decision to negotiate a “zero yield” comprehensive test ban: “As a part of this arrangement, I am today directing the establishment of a new annual reporting and certification requirement that will ensure our nuclear weapons remain safe and reliable under a comprehensive test ban.” For more on the Annual Assessment process, see Tyler (2001).
• The W76 is undergoing a thorough LEP, with the first production unit having been certified prior to this writing. This LEP stands as an excellent example of both the difficulty of extensive refurbishment (see Box 1-2) and the impressive ability of the SSP to respond to technical challenges. The refurbished primaries include parts newly manufactured to original design specifications. Production of “Fogbank”4 was initially problematic but appears to have been achieved satisfactorily.
• There have been a number of technical and programmatic initiatives affecting the nuclear weapons stockpile that have not been implemented for a variety of reasons. These include the Robust Nuclear Earth Penetrator (RNEP), the Reliable Replacement Warhead (RRW), a broad-scale complex transformation plan, and efforts to recapitalize specific elements of the production complex. In some cases, these did not happen because of technical or political objections; in other cases, they failed for lack of policy guidance; and in many situations, there were not sufficient resources to execute them.
• Production facilities have continued to age. Of particular concern are facilities in the Y-12 National Security Complex in Oak Ridge, Tennessee, which conducts uranium processing, and the Chemistry and Metallurgy Research (CMR) Facility at LANL, which provides support for both pit production and plutonium research. These facilities are old and at risk of being shut down from time to time for safety reasons. Their replacement has been delayed due to budget pressures and the fact that the appropriate direction, scale, and scope of the future nuclear weapons program was not yet settled. The FY 2011 budget proposes full funding for both the Uranium Processing Facility (UPF) at Y-12 and the Chemistry and Metallurgy Research Replacement facility (CMRR) at the Los Alamos National Laboratory (LANL) (the FY 2012 request formally deferred CMRR for at least 5 years). Both facilities are also explicitly called for in the 2010 Nuclear Posture Review (U.S. DOD, 2010a, p. 42). In November 2010, the White House announced that the President’s 2012 budget would include additional funding for these two major facilities and stated, “Funding requirements will be reconsidered on an ongoing basis as the designs mature and as more information is known about costs” (Public Law 111-84 Update, 2010, p. 6).
• Two major changes to the surveillance program have begun since the 2002 Report.5
1. The Enhanced Surveillance Program (ESP) was established with the aim of enhancing the non-destructive diagnostic tools available for surveillance.
2. The Surveillance Transformation Project (STP) was created with the goal of collecting data that is more relevant to the needs of today’s stockpile stewardship—for example data that can help quantify aging trends while dismantling fewer units of each warhead type each year.
Although the goals and ideas behind ESP and STP are appropriate, there has not yet been adequate follow-through on budgetary resources and detailed plans needed for implementation.6
• LANL and LLNL have continued plutonium (Pu) aging work and have concluded that, although no critical aging issue had been identified, further work on certain aging topics
4 Fogbank is a material used in the W76 warhead. It was manufactured during the 1980s, and after production ceased in the mid-1990s, substantial effort was required to recertify the production process. This recertification was completed in 2008. For additional details, see LANL (2009); see also: http://www.lanl.gov/orgs/padwp/pdfs/nwj2_09.pdf.
5 Surveillance funding declined from $195 million in FY07 to $158 million in FY09. The FY11 budget increases this amount to $239 million, which the White House has committed to sustain for the next several years. Public Law 111-84 Update, p. 3
6 See also the discussion of surveillance on p. 23.
(e.g., phase stability, engineering behavior, hostile environments, and corrosion) is warranted to reduce uncertainties, explore whether there are any further materials issues, and develop any associated mitigation strategies.
• A shift has taken place in the way in which the NNSA and the Laboratories formally analyze stockpile systems. In the 2002 Report, the key element in evaluating (or enhancing) warhead performance was primary margin (the degree to which the primary yield exceeded the minimum level needed to provide the required secondary yield). Since that time, this concept has been extended to other aspects of a weapon’s performance and is now practiced within a framework known as Quantification of Margins and Uncertainties (QMU).7 Thus, for example, when assessing a proposed LEP, the QMU analyses provide a way of quantitatively comparing various options in an objective, systematic, and transparent way. QMU also provides a metric for the stewardship program, whose goal is to use the various scientific tools to reduce uncertainties and thus increase confidence in the ability to predict performance, safety, security, and reliability in an existing or modified weapon.
• A significant shift in the perspective on stockpile management has occurred since the 2002 Report. The conclusion in 2002 was that the nuclear stockpile could be maintained in the absence of nuclear-explosion testing and that changes, especially to the nuclear explosive components, were high-risk and should be avoided. Since that time, there have been improvements in technical understanding as well as an increased emphasis on nuclear warhead safety and security. As a consequence, the United States is now considering as serious options changes to the nuclear explosive package incorporating additional safety and security features and/or increased performance margins8 in addition to the original LEP approach of refurbishment (see Box 1-2). In particular, the possibility of reuse of nuclear components from other tested systems (other than the one under maintenance) and/or nuclear component design changes (within the range of U.S.-tested designs) that improve the safety, security, and reliability could be credible and technically feasible approaches. We caution that certification of some safety and security improvements under consideration would require detailed analysis on a case-by-case basis and, in some cases, greater scientific understanding than now exists. The 2010 Nuclear Posture Review (NPR) states that the United States will “study options for ensuring the safety, security, and reliability of nuclear warheads on a case-by-case basis…. The full range of LEP approaches will be considered: refurbishment of existing warheads, reuse of nuclear components from different warheads, and replacement of nuclear components.” The NPR goes on to state a “strong preference” for refurbishment or reuse, with replacement considered only where critical goals could not otherwise be met (U.S. DOD, 2010a, p. 39).
Finally, in addition to releasing its Nuclear Posture Review, the Administration has concluded a “New START” Treaty with Russia and ratified it on February 2, 2011. The follow-on to the expired START I Treaty between the United States and the Russian Federation will reduce deployed strategic nuclear warheads on each side to 1,550. The Nuclear Posture Review, the third since the end of the Cold War and first to be unclassified, includes new
7 QMU requires the quantitative (“Q”) assessment of performance in terms of margins (“M”—the difference between the best-estimate value and the threshold value for key performance characteristics) and uncertainties (“U”—the amount by which the best-estimate and threshold values are uncertain). When the ratio of M to U is well above one, then there is high confidence that performance will be acceptable, while an M/U ratio closer to or less than one indicates lower confidence. See NRC (2008).
8 For some weapons, higher margins are key to confidence in performance under certain circumstances, if they can be achieved without an accompanying increase in uncertainties.
guidelines for U.S. nuclear weapons policy. The report outlines five key objectives for U.S. nuclear posture (U.S. DOD, 2010a, p. 2.):
1. Preventing nuclear proliferation and nuclear terrorism.
2. Reducing the role of U.S. nuclear weapons in U.S. national security strategy.
3. Maintaining strategic deterrence and stability at reduced nuclear force levels.
4. Strengthening regional deterrence and reassuring U.S. allies and partners.
5. Sustaining a safe, secure, and effective nuclear arsenal.
Elaborating on the fifth objective, the NPR states that, as long as nuclear weapons exist, the United States will sustain safe, secure, and effective nuclear forces, which will continue to play an essential role in deterring potential adversaries and reassuring allies and partners around the world.
BOX 1-2 Options for Extending the Life of Nuclear Warheads
Under any life-extension plan for a warhead, components outside of the nuclear explosive package (NEP) have been, and will be, replaced with upgraded versions—this is not controversial. For the NEP itself, the terms “refurbishment,” “reuse,” and “replacement” are frequently used in the discussion of options for life-extension programs. However, their definitions are not always clear and are not universal. Below, the committee defines these terms as used in this report:
• Refurbishment describes the case in which individual components in the NEP are either retained for continued use or replaced with components of nearly identical form, fit, and function.
• Reuse describes the case in which pits and secondary components from different, previously fielded warhead designs are introduced into the warhead. This usually implies that the pits and/or components are taken from existing surplus stocks, but if such parts did not exist in sufficient number, the committee would extend “reuse” to include parts newly manufactured to nearly identical specifications.
• Replacement describes the case in which pits and/or secondary components introduced into the warhead are based upon previously tested designs but may differ in some respects from such designs.
Finding 1-1: The technical capabilities for maintaining the U.S. stockpile absent nuclear-explosion testing are better now than anticipated by the 2002 Report.
Current Status of U.S. Nuclear Weapons and the Stockpile Stewardship Program
The number of deployed U.S. nuclear weapons has continued to decrease (through the vehicle of the Strategic Offensive Reduction Treaty [SORT], also called the Moscow Treaty of 2002) and, as of December 31, 2009, stood at 1968 (U.S. Department of State, 2010a) operationally deployed strategic warheads. The “New START” Treaty signed in Prague on April 8, 2010 (and ratified by the United States on February 2, 2011) by the United States and the Russian Federation reduces deployed strategic nuclear warheads on each side to 1,550, down from the 2,200 limit set by the Moscow Treaty (U.S. Department of State, 2010b and Public Law 111-84, 2010).
The total stockpile has also decreased. As of September 30, 2009, the U.S. stockpile of nuclear weapons consisted of 5,113 warheads, a 75-percent reduction from the number in the stockpile when the Berlin Wall fell in late 1989, and a 50-percent reduction from the stockpile existing at the time of the 2002 Report. Several thousand additional nuclear weapons are currently retired and awaiting dismantlement (U.S. DOD, 2010b).
The stockpile consists of nuclear explosives that can be mounted into one or more of three types of delivery systems: aircraft, land-based missiles, and sea-based missiles. Traditional terminology divides the stockpile into strategic and tactical components.9
• Strategic forces include: Bombers (carrying gravity bombs and air-launched cruise missiles, ALCMs), land-based missiles (intercontinental ballistic missiles, ICBMs), and submarine-launched ballistic missiles, SLBMs), together forming the so-called “traditional nuclear triad.”
• The so-called tactical component includes gravity bombs for deployment on U.S. and NATO tactical aircraft, as well as a small number of non-deployed submarine-launched cruise missiles (SLCMs).10
The United States has seven main nuclear warhead designs in its active stockpile today:11 B61 (tactical or strategic bomb), B83 (strategic bomb), W76 (SLBM warhead), W78 (ICBM warhead), W80 (ALCM/SLCM warhead), W87 (ICBM warhead), and W88 (SLBM warhead). Each of these designs was certified before it was introduced into the stockpile and is maintained in operational status with a team examining and evaluating it to support the annual assessment of the stockpile.
Stockpile Stewardship Program Components
The SSP has a broad mandate for maintenance, evaluation, and improvement of the stockpile. Carrying out this mandate requires a broad range of facilities and capabilities:
• Surveillance (including dismantlement, maintenance, refurbishment and assembly of nuclear weapons), centered at Pantex and Y-12;
• Experimental research, primarily located at LANL, LLNL, Sandia National Laboratories (SNL), and the Nevada Test Site (NTS);12
• Design, modeling and simulation, including the Advanced Simulation and Computing (ASC) program at the DOE nuclear weapons laboratories; and
9 Although the difference between a tactical and strategic weapon can be somewhat arbitrary, defined by treaty language or delivery capability, and all nuclear weapons can be viewed as strategic in the sense that they alter the nature of the conflict, this report will continue to distinguish between and utilize the terms strategic and tactical. In this report, a “new type” of strategic weapon refers to a strategic weapon whose design falls outside of the range of those in a country’s nuclear-explosion test experience. See further discussion of the distinction in Chapter 4 and in Box 4-2.
10 Cruise missiles (Tomahawk Land Attack Missiles/Nuclear, TLAM/N) were removed from U.S. Navy vessels as a result of the President’s 1991 Nuclear Initiatives, but the United States has the option to redeploy them on attack submarines if it decides that action is necessary. The 2010 Nuclear Posture Review eliminates TLAM/N.
11 B designates bomb; W designates warhead. There are multiple modified forms of some of these designs, especially the B61.
12 The NTS was recently renamed the Nevada National Security Site (NNSS); however, the acronym NTS will continue to be used in this report due to its familiarity to most readers.
• Nuclear weapons production at LANL, Pantex, and Y-12, for components involving special nuclear material; Savannah River Site (SRS) for tritium production; and SNL and the Kansas City Plant for non-nuclear components.
The DOE nuclear weapons laboratory directors and the STRATCOM Commander rely on the SSP to provide the information underpinning their conclusions in the annual assessments of the nuclear weapons stockpile.
The Surveillance Program pulls deployed warheads from the stockpile and conducts non-invasive testing (e.g., x-ray) on some and invasive testing (disassembly) on others. Both nuclear and non-nuclear components are examined. A vigorous surveillance program is an essential mainstay of the nuclear weapons effort because it allows NNSA to examine warheads for anticipated degradation from known age-related processes and for unanticipated problems related to age or production defects. The United States initiated its stockpile surveillance and assessment program in 1958 when sealed pits were introduced into stockpile weapons. The original surveillance program was oriented primarily toward finding production defects. The number of warheads of each type to be inspected each year was chosen so that if a particular defect occurred in the warheads, there would be a high probability of inspecting a defective unit within a two-year interval. In recent years, it has been proposed that this original program be replaced with the Surveillance Transformation Project (STP), which applies a more targeted data collection to fewer units.
Assessments to date have been made with acceptable uncertainties, but assessment uncertainties will increase over time unless the surveillance program begins to provide the needed quality and type of new data (Congressional Commission, 2009; JASON, 2009). The committee learned that the surveillance program for the past several years has not been providing the timely data that the warhead design teams will need for confident assessments of aging and other effects into the future. Because surveillance is so central to successful stewardship, the committee is gravely concerned about any possible shortcomings in the activities documenting the health of the stockpile. NNSA has advised the committee that the President’s FY 2011 budget will rectify resource shortfalls. The committee has not evaluated the NNSA plan and is principally concerned with long-term sustainment of the necessary efforts.13 However, in a letter of December 1, 2010, to Senators Kerry and Lugar, the directors of the 3 weapon laboratories state that “… the proposed budgets provide adequate support to sustain the safety, security, reliability and effectiveness of America’s nuclear deterrent within the limit of 1550 deployed strategic warheads established by the New START Treaty with adequate confidence and acceptable risk” (Miller et al., 2010).
Finding 1-2: Future assessments of aging effects and other issues will require quantities and types of data that have not been provided by the surveillance program in recent years.
Experimental Research Facilities
The SSP has a number of large-scale experimental facilities that to some extent simulate phenomena that occur during a nuclear explosion. During the past several years major
13 For White House commitments to maintain adequate surveillance funding; see Public Law 111-84 Update, p. 3.
advances have been made in these facilities, and they have become integral parts of the present and future stewardship program. These include:
National Ignition Facility (NIF). NIF was dedicated on May 28, 2009, at LLNL. Its goal is to provide a thorough scientific understanding of the behavior of materials during conditions similar to those in a nuclear explosion, and to carry out both ignition and weapons physics experiments to study dynamic phenomena that occur during such explosions. A major objective is to achieve ignition (i.e., thermonuclear burn) because this will enhance the U.S. ability to study such phenomena. The early experiments since the NIF dedication have demonstrated that the facility is operating well. All 192 beams have been fired a number of times and total power levels on target have reached the one megajoule (MJ) level (of the ultimately planned 1.8 MJ, or 3 MJ of green light), with many of the issues important for ignition already being tested. The current emphasis is on executing the National Ignition Campaign (NIC) with the goal of achieving ignition within the next two years. Throughout this period, there will also be a major effort aimed at detailed comparison of the simulation codes with experiments. Subsequent to the NIC, the facility will broaden its program to act as a national user facility with a portion of the experimental time made available to the general scientific community. Concurrently, the NIF will be instrumental in making progress on the major weapons initiatives in boost and energy balance, as well as testing many aspects of weapons design codes.
• Dual-Axis Radiographic Hydrodynamic Test (DARHT) Facility. This facility at LANL is a complete radiographic system with elements that combine to produce multiple images of test objects as they evolve during an implosion, with unprecedented resolution. Images from the dual axes allow scientists to check the implosion symmetry of weapon mockups because images can be taken from orthogonal directions at the same time. The first axis has been operational since 1999, and the first hydrodynamics experiment using both axes was successfully executed on December 3, 2009.
Image data and data from other measurements made during the hydrodynamic tests are compared with computer calculations to assess the ability to predict the performance of systems that are similar to real weapon systems. Scientists at DARHT can now follow the implosion’s progress14 until almost the explosion time of a real weapon and compare the pictured component positions with the predictions of computer simulations. In addition, DARHT can be used to study basic weapons physics with scaled experiments and to develop a deeper understanding of detonations, hydrodynamic behavior, and materials properties. This information is then incorporated into new simulation computer codes developed under the Advanced Strategic Computing program.
• Microsystems and Engineering Sciences Application (MESA). MESA was constructed at Sandia National Laboratories to provide advanced simulation tools, microsystems, and nanotechnology capabilities for weapons engineering and related national security projects. It consists of three elements: the Microelectronics Development Laboratory, the Microsystems Laboratory, and the Weapons Integration Facility. It was dedicated on August 23, 2007.
Sandia has the primary role of developing and designing the electronic systems that operate nuclear warheads. Key components of those systems are radiation-hardened microelectronics, devices that are designed, qualified, and fabricated in the MESA facilities. In addition to fabricating electronic circuits, the MESA facility also makes microelectromechanical elements for advanced security systems, sensors, guidance systems, and other applications. The complex also includes the world’s most complete
14 Using a simulant for weapon-Pu.
compound-semiconductor fabrication facility. This will produce advanced optoelectronic and custom electronic components, communications, and other emerging technologies. It is one of the few microelectronic facilities that is fully integrated with a state-of-the-art high performance computing capability.
Other Facilities. In addition to NIF, DARHT, and MESA, a number of other smaller facilities contribute substantially to the stewardship program.
• The OMEGA laser at the University of Rochester has been important in providing data from high-energy-density experiments and in preparing the technical path for NIF, and it continues to be a complementary facility.
• The Z Machine at Sandia provides a pulsed power capability that enables scientists to probe many relevant topics in high energy density science.
• Both the Joint Actinide Shock Physics Experimental Research (JASPER) Facility and the underground U1a Complex at the Nevada Test Site are dedicated to dynamic experiments on the properties of uranium and plutonium.
• In addition, each of the three laboratories has special-purpose facilities, such as the High Explosives Applications Facility (HEAF) for high explosive work at LLNL, the Los Alamos Neutron Science Center (LANSCE) for neutron experiments, and various machines that are used for work on weapon vulnerability in hostile environments at Sandia and NTS.
Design, Modeling, and Simulation
Advanced Simulation and Computing (ASC)—Even during the nuclear weapon testing era, computer simulation of nuclear weapons performance was the primary way in which nuclear weapons were designed. In the absence of nuclear-explosion testing, advanced computing serves as the means for putting all of the nuclear weapons information together to evaluate the impact of a surveillance finding, assessing the impact of a proposed refurbishment or modified design, and comparing experiments on facilities such as NIF or DARHT with calculations made with weapons codes (testing the merits of both the code and the designer).
In the early phases of the SSP, the primary goal of the advanced computing program was to establish a high resolution, 3-dimensional (3-D) weapons-simulation capability, because assessing age-related problems or other issues tended to involve “off-center” defects such as corrosion and cracks that could not be modeled in a 2-D calculation. Concurrently, the execution of an LEP required detailed design calculations with sufficient detail to provide confidence in the refurbishment. The long-term goal in the computing program is to improve predictive capability with quantitatively assessed uncertainties so that there will be even higher confidence in proposed warhead modifications for an LEP, including the incorporation of new safety and security or performance features.
Since the mid-1990s, inception of the Advanced Simulation and Computing Initiative (ASCI, the predecessor of ASC), the computing capability available to weapons designers has increased by a factor of approximately one hundred thousand, as shown in Figure 1-1. In recent years the strategy has been to pursue “capability” machines for the most detailed calculations, “capacity” machines for cost-effective high throughput of smaller calculations, and “advanced architecture” machines to explore and influence potential game-changing technologies. Each kind of machine plays an important role in the weapons program. It is possible that the “capability” and “advanced architecture” machines may be one and the same in future acquisitions. A recent program change is to have a single “capability” machine (on which the most detailed and demanding calculations are carried out) accessible by all three Laboratories
on a secure network. This has the virtues of much greater efficiency and reduced cost, while providing a common framework for comparison. The current computer is located at LLNL, with the next one to be sited at LANL, with Sandia participating in the development.
“Capacity” machines—less powerful but less expensive than capability ones—do much of the “get-ready” calculations to prepare for full-design simulations, and a new tri-laboratory architecture is now used by all the laboratories. Advanced architecture machines tend to have great speed but much less flexibility in the kinds of problems on which that speed is effective. Thus, existing computers at LLNL and LANL, and two new computers planned for LLNL in the next two years, can carry out valuable, special-purpose calculations (often focused on basic weapons science issues) while scientists explore their adaptation to direct design work.
FIGURE 1-1: Advances in computation power in the Advanced Simulation and Computing (ASC) program since 1996. SOURCE: NNSA
Exascale computing, the next major threshold, will not be achieved by a natural evolution of existing technologies. Basically, the elements of the modern supercomputer—thousands of individual, interconnected smaller computers—are continuing to gain in speed but not adequately in memory density. At the same, time the power requirements are becoming unsupportable in cost and cooling. These technology barriers will become limiting factors over the next 5 years, before exascale-class systems will be reached. Consequently, multiple hardware and software paths towards exascale are being pursued, some in partnership with other organizations, such as DOE’s Office of Science.
Nuclear Weapons Production Complex
The nuclear weapons production complex includes a number of sites that deal with different elements of weapons systems:
• Pantex Plant (high explosives, assembly and disassembly of weapons);
• Y-12 National Security Complex (uranium and other components);
• LANL (detonators and plutonium activities including pit production);
• Savannah River Site (tritium production and gas transfer systems);
• SNL (neutron generators); and
• Kansas City Plant (electronic, electrical, and mechanical components, including gas transfer systems).
Immediate needs include modern facilities that support uranium and plutonium science, and provide capabilities to produce pits and uranium components for stockpile weapons. The two sites responsible for production activities with special nuclear materials, Y-12 and the plutonium complex at LANL, rely on facilities that in some cases date from the early days of the Cold War, and it is both expensive and difficult to maintain them at modern safety and security standards. As was mentioned earlier in this chapter, NNSA has detailed modernization plans to build replacement facilities, but replacement of the main facilities has been delayed due to budget pressures and the fact that the appropriate direction, scale, and scope of the future nuclear weapons program was not yet settled. DOE and DOD officials have testified that the United States maintains a large number of reserve warheads because of the inability to manufacture significant numbers of new warheads in response to changed geopolitical circumstances.15
• Y-12. This site is responsible for the manufacture of secondaries, including uranium components. NNSA’s plan is to consolidate the storage areas for highly enriched uranium (HEU) at the HEU Materials Facility (HEUMF), which has been built, and to build a uranium processing facility (UPF) to replace several 50-year-old facilities. The projected cost for this plan is estimated as $4.2-6.5 billion16 to create a production capacity of 80 units per year. The benefits of the new Y-12 facility would be a smaller footprint, leading to much lower security costs, a safer facility, and one that is appropriate for a modern stockpile size (compared with the production requirements of 1,000 per year during the Cold War). However, the lead time to develop an operational new plant is a decade or more, so the overall strategy to ensure the supply of secondaries as needed for LEPs is still under active discussion.
• Chemistry and Metallurgy Research Facility (CMR). LANL facilities have long-term responsibility for the plutonium research and development and pit production work. CMR was completed in 1952 and supports a broad range of plutonium research and development activities as well as pit production. It is recurrently on the verge of being shut down because of safety issues, and there is a plan to replace it with a modern set of facilities, the CMR Replacement complex (CMRR). As with Y-12, the timescale
15 The 2010 Nuclear Posture Review discusses the need for non-deployed warheads as a “hedge against technical or geopolitical surprise” and notes that there are currently more warheads than are required. It further states that “progress in restoring NNSA’s production infrastructure will allow those excess warheads to be retired” (U.S. Department of Defense, 2010a, p. 38).
16 Public Law 111-84 Update, p. 6.
to complete the facility is on the order of a decade. The total project costs are estimated at $3.7–5.8 billion.17
• Pit Production Facility (PF-4)/Technical Area-55. The LANL plutonium production facility (PF-4) is the only one in the United States that can produce war-reserve quality plutonium pits for the stockpile. Recently, LANL has demonstrated its capability through production of a certified W88 pit. Current capacity at LANL is stated to be 6-10 war-reserve pits per year. With major infrastructure investments, including completion of CMRR, the maximum war-reserve production rate of 80 pits/year could be achieved. Relatively modest investments in upgrades to PF-4 could result in a pit reuse capacity of at least 40 war-reserve pits per year.
The President’s FY 2011 budget requested funding for construction of both the UPF at Y-12 and CMRR at LANL; the FY 2012 request formally deferred CMRR for at least 5 years.
Life-Extension Programs (LEP)
Since the 2002 Report, there has been considerable discussion about the changes that accumulate over a warhead’s lifetime. In such discussions, four distinct categories of changes should be kept clearly separated:
1. Changes induced by aging;
2. Changes from the certified design introduced during manufacturing;
3. Changes in the assessed performance of the certified design driven by improved understanding or more careful assessment; and 6.
4. Deliberate physical changes (made through an LEP or another program of alteration).
Changes in all four categories have taken place in the stockpile. The appropriate response depends on the category and other details. When an aging-induced change is assessed to reduce confidence to an unacceptably low level, a deliberate physical change is required to restore the warhead’s original performance characteristics. However, aging-induced changes do not argue for one kind of LEP over another, because a LEP of any kind can reset the “aging clock” to zero. Discovery of a manufacturing error prompts an assessment of how many units it may affect and the creation of an acceptable solution. Changes in the assessed performance of a design could lead to modification of certified performance characteristics, modification of change-out intervals for limited-life components, or a decision to remove the weapon from the stockpile. Deliberate physical changes are made when necessary and are made in the course of LEPs, which take place for a given NEP only once every few decades.
The committee investigated concerns that have been expressed about accumulated changes in the stockpile from aging and other sources. In probing to understand these concerns, the committee was impressed by the degree to which technical issues encountered to date have been resolved by the U.S. nuclear complex. In their discussion with the committee, the laboratory directors from Livermore, Los Alamos, and Sandia all indicated that there is no evidence of any technical issues that cannot be resolved with the present competency (LLNL, October 2009, personal communication). In their annual assessment letters from 2009, the laboratory directors all indicate that underground nuclear-explosion testing is not presently required to maintain the certification of weapons in the stockpile. They expressed concerns, however, about financial and other constraints on the technical program, deferred life extension decisions and activities, and shortfalls and delays in surveillance data. They warn that aging and
other accumulated changes to weapons in the stockpile will erode confidence in weapon performance in the future unless surveillance programs are enhanced and given priority, LEPs are conducted in a timely way, and investments in the physical and intellectual infrastructure of the nuclear weapons enterprise are increased.
The committee offers several observations on concerns about aging and accumulated changes. First, each annual assessment letter stated that nuclear-explosion testing was not necessary at the time of its writing to resolve technical issues. Second, changes caused by aging should be discussed separately from other categories of changes. The existence of age-induced changes is not a surprise. We have long known that they would occur and that, if unaddressed, “continuing accumulation of aging changes” would reduce reliability and effectiveness. However, they can be eliminated through any of the LEP options that have been considered, if funded and performed as needed. Third, it is easy to misinterpret the phrase “accumulated effects of weapons changes” as referring to something that is in addition to aging-induced changes. In fact, aging-induced changes are the only physical changes that are accumulating year to year in the nuclear explosive packages (NEPs) in the stockpile. Deliberate physical changes are carefully designed to have at most a minimal negative impact on confidence, and in many cases are assessed to improve confidence. Fourth, manufacturing errors should be discussed separately from changes in assessed performance of a given design, for they have different causes and different solutions.
Finding 1-3: The committee judges that Life-Extension Programs (LEPs) have been, and continue to be, satisfactorily carried out to extend the lifetime of existing warheads without the need for nuclear-explosion tests. In addition to the original LEP approach of refurbishment, sufficient technical progress has been made since the 2002 Report that re-use or replacement of nuclear components can be considered as options for improving safety and security of the warheads. The assessment of which of the spectrum of options is most appropriate for a given warhead needs to be made on a case-by-case basis because the benefits and risks depend on the warhead under consideration. Replacement of non-nuclear components has always been an option.
The workforce and management of the nuclear weapons complex are essential elements of maintaining the safety, security, and reliability of the U.S. stockpile. To sustain a technically competent, motivated and capable workforce, the weapons laboratories will need to employ some degree of innovation, for example by developing a work structure that draws from a broader base of nuclear-capable personnel to work on more diverse national security projects. To be successful, such activities must be encouraged and supported by NNSA and laboratory management as well as Congress and the Administration. These topics are addressed further in Chapter 3.
The discussion in the above sections leads to the following finding and recommendation:
Finding 1-4: Provided that sufficient resources and a national commitment to stockpile stewardship are in place, the committee judges that the United States has the technical capabilities to maintain a safe, secure, and reliable stockpile of nuclear weapons into the foreseeable future without nuclear-explosion testing. Sustaining these technical capabilities will require at least the following:
• A Strong Scientific and Engineering Base. There must be continued adherence to the principle that the ability to assess and certify weapons rests on technical
understanding of weapons phenomena, data from past nuclear-explosion tests, computations, and data from past and ongoing experiments. Maintaining both a strategic computing capability and modern non-nuclear-explosion testing facilities (for hydrodynamic testing, radiography, material equation-of-state measurements, high explosives testing, and fusion experiments) is essential for this purpose.
• A Vigorous Surveillance Program. An intensive surveillance program aimed at discovering warhead problems is crucial to the health of the stockpile.
• Adequate Ratio of Margin to Uncertainty. Performance margins that are sufficiently high, relative to uncertainties, are key ingredients of confidence in weapons performance.18
• Modernized Production Facilities. Most of the nuclear weapons production facilities are old (50 years in some cases) and are both difficult and costly to operate in accordance with modern standards of safety and security.
• A Competent and Capable Workforce. Nuclear weapons work (e.g., the SSP) is key to meeting a range of challenges in the broader national security landscape. Exploration of these broader areas (e.g., nonproliferation programs, render safe, etc.) can provide opportunities for intellectual stimulation and professional development that will attract a diverse, capable workforce. It is equally important to ensure that the Department of Defense, particularly the Defense Threat Reduction Agency, the Navy’s Strategic Systems Project Office, and the Air Force’s Ballistic Missile Organization maintains a technically competent workforce (Congressional Commission, 2009; U.S. Secretary of Defense Task Force on DOD Nuclear Weapons Management, 2008; Defense Science Board, 2008).
Recommendation 1-1: To address each of the essential elements of stockpile stewardship listed in Finding 1-4, NNSA, working with the Administration and Congress as appropriate, should:
• Maintain a continuing dynamic of experiments linked with analysis. Both are essential to maintaining the capability to render judgments about stockpile issues.
• Maintain a vigorous surveillance program that is systematic; is statistically based where possible; and continuously reflects lessons learned from annual surveillance, LEPs, fixing problems, and science-based analysis. Nondestructive tools and experimentally validated computational analysis should be developed and applied to introduce more predictive capability into the surveillance system.
• As part of each LEP, explore options for achieving adequate margins through reuse or replacement scenarios in addition to refurbishment, to determine how best to meet military, technical, and policy objectives. Assess uncertainties associated with each scenario.
• Develop and implement a long-term production facility modernization plan. This should include maintaining a plutonium science and production capability, including the ability to produce various types of pits for weapons in the stockpile.
• Broaden the base of its nuclear expertise by involving nuclear-capable personnel in related national security projects (nuclear forensics, intelligence, threat reduction programs, basic science applications of stewardship activities, etc.).
18 Some of today’s systems already have relatively high margin-to-uncertainty ratios; others are relatively low.
The 2002 Report did not evaluate test readiness in any detail. Since that report was issued, there have been a number of developments.
During the 1999 consideration of CTBT ratification, the Administration proposed the following as a safeguard: “The maintenance of the basic capability to resume nuclear test activities prohibited by the CTBT should the United States cease to be bound to adhere to this Treaty.” The committee presumes (and would favor) inclusion of such a safeguard when CTBT is reconsidered. Underground nuclear-explosion testing requires a suitable test site, a set of specialized equipment and infrastructure, and a body of specialized knowledge. As long as the NTS remains in government possession and free from any construction that would interfere with nuclear-explosion testing, reconstituting a “basic capability” will, in principle, always be feasible. The question is, therefore, how much lead time the United States should assume would be necessary for a resumption of testing.
Currently, NNSA is required by a 1993 Presidential Decision Directive (PDD-15, “Stockpile Stewardship”) to maintain the ability to conduct a nuclear test within 24-to-36 months of direction by the President to do so.19 This is the only extant guidance on the timing of test readiness. The 2010 Nuclear Posture Review is silent on the subject of test readiness. The 2001 Nuclear Posture Review called for shortening this period. Although the Administration did not specify a time, NNSA internally adopted a goal of being ready to conduct a test in 18 months (Brooks, 2005). Congress declined to fund 18-month readiness and directed (in report language accompanying the FY 2006 Energy and Water Development Appropriations Act) that the Administration maintain the ability to conduct an underground nuclear-explosion test within 24 months. Since that time, funding reductions by Congress and competing priorities within NNSA have further reduced funding for test readiness.
From the technical perspective, there are four major reasons why a limited number of tests might be required:20
• To develop a weapon with new technical characteristics that could be validated only by testing.
• To confirm a stockpile problem or certify a solution.
• To ensure understanding of actions taken by others (technical surprise).
• To verify render-safe procedures.
NNSA currently assesses that it has the capability to conduct a test to meet very limited technical objectives within 18 months of a request to do so, but only if some “domestic regulations, agreements and laws” were to be waived.21 With greater confidence, NNSA assesses that it could conduct a test and achieve specific technical objectives within 36 months
19 Discussion of required lead times is based on Nuclear Test Readiness: Report to Congress (U.S. DOE, 2009). Note that NNSA assessments assume the conduct of only a single test or very short series; NNSA is not maintaining any capability to resume routine, sustained nuclear testing.
20 Some would add weapons effects tests; the committee judges these as less significant. In any case, such tests are sufficiently complex that designing them is likely to be the pacing item in preparing for a test. A final possibility is a test to demonstrate resolve or one conducted in response to testing by another State. In such a case, the United States would presumably wish to conduct the test as rapidly as possible but would have no specific technical objectives. The committee considers such testing outside the scope of its charter.
21 “Currently, the NNSA does not have the capability to conduct a nuclear test that would be fully compliant with domestic regulations and laws with[in] 18 months of an order by the President to do so” (U.S. DOE, 2009).
of a decision. However, because of technical uncertainty, NNSA cannot say with “high confidence that [it] remains within the required 24-36-month required time window for highly diagnosed and authorization-driven hypothetical nuclear tests.” The committee also heard the view that investing annually in maintaining aging test diagnostic equipment and other test-related capabilities may not be the best possible use of funds. If events led to a decision by the United States to resume testing, presumably the crisis would be severe enough to justify the identification of funding specifically to support testing. Such funding would allow for the procurement of the best available new equipment. The pacing item in conducting an underground test is likely to be regulatory, not technical. For example, the National Environmental Policy Act (NEPA) requires extensive analysis that is often subject to court challenge.22 Similarly, Title 10 of the Code of Federal Regulations (CFR) Part 830, Nuclear Safety Management, requires NNSA to complete various safety analyses before conducting a test. In addition, Congressional funding approval would be required. Some—and perhaps all—of these regulatory requirements could be waived by the President (or by legislation) if the situation were considered urgent enough to do so.23 Assuming waivers of regulations and laws, the only legal limit on how rapidly the United States could conduct a test would be the obligation in Section IV of the Protocol to the Threshold Test Ban Treaty of 1974 to notify the Russian Federation 200 days in advance of a nuclear-explosion test.
Development of a weapon with new military characteristics would take significantly longer than 24-36 months. Thus the important limit appears to be the need, however remote, to conduct a test to ensure the health of a weapon important to the stockpile. The pre-2006 NNSA readiness goal of 18 months was based on analysis indicating that, when the United States was routinely testing prior to 1993, it would normally take about 18 months to develop and field a nuclear-explosion test designed to obtain technical data (U.S. DOE, 2009).
For nearly a decade, Congress has consistently reduced or denied funding for test readiness. NNSA elected to seek no funding for Fiscal Year 2010, and the President’s fiscal year 2011 budget has no dedicated funding for test readiness. As a result, NTS is experiencing many of the same problems that are seen in other elements of the nuclear weapons complex, including age-related degradation of physical assets and diagnostic equipment, lack of maintenance, outdated technology, and lack of experienced and trained personnel for critical positions. A recent report by the Department of Energy Inspector General concluded that, “there is a risk that physical assets and diagnostic equipment could not be made ready to support an underground nuclear test within the required three-year window” (Sedillo, 2009). The Inspector General attributed these problems to lack of budgetary support.
Although SSP provides many of the capabilities required to conduct a nuclear-explosion test, the committee believes that continued use of the NTS for sub-critical experiments and for other complex, high-hazard operations also helps maintain test readiness. It is important to note that some capabilities must be explicitly maintained in addition to a robust SSP. NNSA identified the following in presentations to the committee (U.S. DOE, 2009):
• Preserve the Nevada Test Site’s ability to host a nuclear-explosion test;
• Support containment capability unique to underground nuclear-explosion testing;
22 In the specific case of NEPA, Title 40 of the Code of Federal Regulations (CFR) Part 1506.11, Emergencies, authorizes alternate arrangements with the approval of the White House Council on Environmental Quality in cases “where emergency circumstances make it necessary to take an action with significant environmental impact without observing the provisions of these regulations.”
23 Recently a task force, “which included all three NNSA national laboratory directors, concluded” that “a very limited test to signal the readiness of the U.S. nuclear deterrent or respond to another nation’s test, could be conducted in 6 to 10 months, but such a test is not a component of stockpile stewardship” (U.S. DOE, 2011).
• Maintain the integrity of the seismic analysis of southern Nevada;24
• Maintain the test-specific parts of radiochemistry infrastructure and drill-back capability;
• Support fast readout requirements; and
• Maintain a library of testing methods, rack designs, procedures, processes, etc.25
Finding 1-5: To maintain a test readiness capability of 24-36 months as required by PDD-15, some test readiness capabilities must be explicitly maintained in addition to the Stockpile Stewardship Program. Test readiness draws on SSP capabilities but in addition requires a suitable test site, a set of specialized equipment and infrastructure, and a body of specialized knowledge. The pacing item in resuming nuclear explosive testing may be regulatory rather than technical.
Recommendation 1-2: To maintain a test readiness posture of 24-36 months, NNSA should:
• Preserve the Nevada Test Site’s ability to host a nuclear-explosion test;
• Support the containment capability unique to underground nuclear-explosion testing;
• Maintain seismic data necessary to meet U.S. obligations under the Threshold Test Ban Treaty should testing resume;
• Maintain the radiochemistry laboratory infrastructure and drill-back capability;26
• Support fast readout requirements and prompt diagnostic equipment;
• Maintain a library that includes testing methods, containment rack designs, procedures, processes; and other relevant information;
• Maintain nuclear-certifiable emplacement cranes;
• Maintain field-test neutron generators;
• Establish a process for obtaining waivers from health and environmental regulations if required, but, given the frequency with which laws change, not seek such waivers in advance.
NNSA should include all of these elements within the SSP and evaluate their status as part of the annual assessment of the fulfillment of safeguards recommended in Chapter 3 of this report.
24 This refers to maintaining current data required to be provided to the Verifying Party under the Protocol to the 1974 Threshold Test Ban Treaty (TTBT). Were the United States to resume testing, Russia would have inspection rights under the TTBT. Depending on the specific test, paragraph 9 of Section IV of the Protocol would require providing “a description of the geological and geophysical characteristics of the test location, which shall include: the depth of the water table; the stratigraphic column, including the lithologic description of each formation; the estimated physical parameters of the rock, including bulk density, grain density, compressional velocity, porosity, and total water content; and information on any known geophysical discontinuities in the media within a radius of 300 meters of the planned emplacement point of each explosive canister” (U.S. Department of State, 1974).
25 The committee has noted the problems stemming from loss of institutional memory in regard to the manufacture of Fogbank (see footnote 11). This should not be repeated in other fields.
26 The U.S Administration has asked the National Research Council to address workforce issues in the area of nuclear chemistry (NRC, 2012, forthcoming).
Specialized technical requirements with respect to the stockpile also exist within DOD. Strategic planners must be well versed in the technical realities of weapons performance to be able to integrate knowledge of the time scales, costs and performance tradeoffs into defining weapons requirements. In addition, the DOD maintains direct technical responsibility for handling nuclear warheads and for the interface of the warheads to the control systems of the various delivery vehicles. An example of the demanding nature of this activity is the Mk-21 fuze required for the W87 warhead in the Mk-21 re-entry vehicle on the Minuteman ICBM. This fuze has a 10-year design life and thus must be replaced routinely as part of LEP activities. However, it is a custom design with rigorous performance specifications and thus is manufactured under direct Air Force control. Failure to maintain the technical knowledge base for this remanufacture has recently resulted in a problem. Addressing this problem has imposed costs and delays that could have been avoided with investment in maintaining the nuclear workforce. The knowledge and skills required for these activities are specialized, and impose the same need for care in workforce morale and sustainment as discussed above in the context of the nuclear complex (Defense Science Board, 2008).
This page intentionally left blank.