Stockpile Maintenance, Manufacture, and Manufacturability. Sciencebased stockpile stewardship involves more than developing the tools to predict performance. As a steward of the stockpile, Los Alamos is also responsible for maintaining the existing stockpile through a program of surveillance and response - taking weapons out of the stockpile, examining them, and solving any observed problems. One type of response is the life extension program. In the next decade, this program will call for replacements or modifications of specific components in the stockpile, and thus it presents major engineering and resource challenges.

Los Alamos has also taken on some production responsibilities as facilities were shut down across the national weapons complex. Our most visible new task is to manufacture the plutonium pit, the heart of the weapon primary, but we are also responsible for manufacturing detonators, neutron generators, beryllium components, and other parts.

In pit manufacture, we have had to recreate the entire technology of the Colorado Rocky Flats Plant in a changed environment, where many materials and processes used at Rocky Flats are neither available nor permitted. Developing and qualifying the new processes and certifying the performance of the product without full-scale testing have been the first big test of the stewardship regime. We have changed not only our technology but also our traditional ways of doing business. Fortunately, our dedicated staff at the plutonium facility responded with their full measure of skill and intensity. By the start of the calendar year, they had produced a number of system qualification test pits and just recently delivered a completely weapons-qualified ("certifiable") pit - a major achievement. In a parallel effort, our program leaders have initiated the development of sophisticated process monitoring and control procedures that guarantee quality during the manufacturing process. This investment in yet another aspect of predictive capability should enable us to sustain the pit manufacturing capability in the present environment of changing requirements and small throughput. Both the life extension program and our different production tasks clearly call for a science-based methodology to establish priorities and quantify our level of confidence in the new or changed components. Responding to an aging component with a plan to replace all identical components in the stockpile and thus "rejuvenate" the stockpile may be a very expensive decision. Without careful assessment of performance versus impact, one can make poor decisions. As described in the next section, we are currently developing a quantitative framework for guiding such decisions and building confidence in stewardship.

A Certification Methodology

    Each year, the director of the Laboratory must assess the weapons in the stockpile for safety, performance, and reliability. This assessment must consider whether military characteristics and requirements can be met without a return to nuclear testing. In the current stewardship regime, the key question we face in the annual certification is, "What is the relationship between key weapon-performance metrics and the design margins of the system?" Furthermore, how far can we stray from the ideal design environment (materials, age, and tolerances) before a weapon will fail to meet its military requirements? And how can we quantify our confidence? That is, how much do we trust our predictions?

These are tough questions that have never before been addressed or quantified. Consequently, the policy community has challenged us to provide a rigorous scientific approach to reach closure on scientific issues and to quantify the level of confidence with which we certify the stockpile. In response, both Los Alamos and Lawrence Livermore have developed a certification methodology that revolves around quantifying margins and uncertainties for the various stages of weapons performance. By judging our progress on the problem of decreasing the uncertainties, we have the means to rank scientific and technical investment. For example, we will be able to decide whether a particular process must be modeled at the molecular or macroscopic level to reduce uncertainty or whether some modest parametric representation would be adequate - all based on assessing the impact of the uncertainty on our confidence in performance. We know that complete predictive capability of weapons performance is not possible, but we will be able to estimate our degree of confidence and specify the requirements for increasing that confidence based on quantitative performance-related measures.

This new methodology has an important corollary. It can help translate the unwritten lore of our best designers into solid guideposts for the emerging generation of new designers. Our best designers, like innovators from every field, did not always write everything down, nor was there ever a prescribed method to document the detailed interplay between simulation and testing. The experienced designers had learned how to compensate for less-than-predictive models by adjusting empirical parameters to ensure enough "predictive ability" in yield and diagnostic measurements and to anticipate the "next" underground test. Now, a new methodology focusing on margins and uncertainties allows for more explicit representation and quantification of essential design decisions and judgment.

Most important in the long term is that certification without testing be sustainable. Sustainable means not only that we continue to increase our science and engineering understanding of the weapon system but that we use that knowledge to make cost-effective decisions about the scope of weapons refurbishment and to better address the issues observed in the aging stockpile.

The Current Global Environment

    Today, the international and national security environments have changed radically and have, to some extent, become entwined. Nations that were once our formidable and determined nuclear enemies have now become our real or emergent allies. Although the Cold War, a struggle that seemed destined to permanence, has ended, the threats to world peace remain real, and arguably, the instability around the globe is greater. Among the new and emergent allies, there is a new determination to stop the growth of this incipient instabilityГone brought to us by the harbingers of terror.

Against such a backdrop, our nation has been reevaluating its nuclear posture. Of course, nuclear capability remains the ultimate deterrent, but ever more voices raise questions about the nature and effectiveness of that deterrent. Here, effectiveness is not discussed in destructive terms, but it refers to maintaining real deterrence against radically different enemies and targets. It may be argued, and it would certainly be ironic, that the existence of nuclear weapons with lower levels of collateral damage and therefore increased "usability" may be the greatest deterrent and thereby the greatest force against their own actual use. The aim would still be to never have to use the weapons.

Policy Changes

    In early 2002, the Bush administration issued the findings from a Nuclear Posture Review that placed nuclear weapons in a new and different context. In the past, we described deterrence in terms of an offensive triad composed of intercontinental ballistic missiles, submarine-launched ballistic missiles, and strategic bombers, each carrying nuclear warheads capable of delivering kilotons, if not megatons, of explosive power. Having evaluated the changed environment in both threat and technology, the Nuclear Posture Review offers a new triad, in which the three nuclear offensive capabilities above appear on one leg of a triangle, joined and complemented by strategic nonnuclear weapons. This change recognizes that precision delivery systems with conventional warheads, such as those exercised during the Gulf War and, more recently, in Afghanistan and Iraq, can now operationally achieve some of the strategic objectives that only nuclear weapons could have achieved in the past.

The first choice is always to avoid direct use of nuclear weapons and to use them only as a deterrent. However, in the event they were required because the destructive effect needed is achievable only through nuclear processes, our nation would not want them to have unacceptable collateral effects. For example, it would be less "effective" to threaten to use a nuclear weapon to destroy chemical and biological agents in a deeply buried and hardened arsenal if the explosion would produce widespread nuclear contamination. Consequently, there may be fewer nuclear weapons in the new triad, but they will probably have to be more robust and address new strategic problems.

The review also introduces a vital, new component to the new triad, namely, responsive infrastructure. In a world where technology is changing quickly, where emerging threats are difficult to identify in advance, the review challenges the science and technology community to develop flexible and adaptive capabilities. What does that mean for the nuclear weapons community?

Advanced Concepts

    In the past, we were asked to build thousands of identical warheads to be placed in ballistic missiles, each directed toward specified targets. Today, the technical and policy communities are increasingly seeing a need for new kinds of devices. Depending on how the threat evolves, we may be tasked to build relatively small numbers of weapons of very special and limited capability. If so tasked, we may extrapolate some of those weapons designs perhaps from the designs in the existing stockpile. Those would be moderately easy to certify without testing. A great number of possible "new" weapons might be based on design concepts and weapons systems that were tested in Nevada before 1992 but never implemented in the stockpile. Depending on the testing pedigree, these may or may not be straightforward to certify without testing.

The Nuclear Posture Review has opened the door to serious thinking about advanced concepts. The timing could not be more opportune. Our experienced designers are nearing retirement, and before they stop working, they must mentor the new designers. Study of advanced concepts offers a dynamic environment for training and transfer of expertise to a new generation. Unlike stewardship of the last decade, which focused on narrow aspects of weapons physics at times, advanced concepts require thinking through the performance of the system as a whole and thus keeping the integrated design capability alive.

The Future and the Need for Talent

    I believe that stewardship is at a crossroads. In the last decade, we have achieved a great deal without testing and have been able to continue to certify the stockpile. However, we are starting to address physics and engineering issues that may not be so amenable to our present tools. For many reasons, the weapons laboratories are not yet able, unfortunately, to develop and validate the new tools fast enough. We have several major stockpile systems to maintain (for example, through life extension), and those efforts are as significant a load as any placed on us during the Cold War.

While the national and international environments compel us to maintain, for the foreseeable future, the science, engineering, and manufacture that underpin the existing nuclear weapons capability, we must also envision how the nuclear community might contribute to a more agile and responsive defense without resorting to testing. In other words, we must create the deterrent of the future.

During the past 10 years, we have prepared for these demanding challenges by embracing a strong scientific approach and developing the tools for sustainable stewardship. Now, we need to continue recruiting and nurturing the best talent to solve the wealth of science and engineering challenges that the program faces. The fact that those problems can now be tackled with some of the most advanced simulation and experimental tools available gives us hope. The determination and continued dedication of our staff sustain that hope.

Article by Raymond Juzaitis in Los Alamos Science, No. 28, 2003. This volume of Los Alamos Science commemorated six decades of service by the Los Alamos National Laboratory.

Raymond Juzaitis received his bachelor's degree in chemical engineering from Princeton University in 1974 and his doctoral degree in nuclear engineering from the University of Virginia in 1980. Ray joined Los Alamos as a doctoral student in the Theoretical Division in 1978. He later became a technical staff member and gradually moved to managerial positions in the weapons program. He is currently associate director for weapons physics and has programmatic responsibility over nuclear weapons physics design and assessment, including Laboratory-wide scientific activities that contribute directly toward the science-based certification of nuclear weapons performance and safety.

(http://www.fas.org/)