The Stockpile Stewardship and Management Program

Maintaining Confidence in the Safety and Reliability of the Enduring U.S. Nuclear Weapon Stockpile

U.S. Department of Energy
Office of Defense Programs

May 1995


Recent changes in national security needs have necessitated corresponding changes in the way the Department of Energy must meet its responsibilities regarding U.S. nuclear weapons. As a result of the START I treaty, the START II agreement, and the recently completed Nuclear Posture Review, the nation's stockpile is being greatly reduced. The Nuclear Posture Review forecasts steady declines in both the size and diversity of the U.S. nuclear stockpile through the year 2003, at which time the strategic accountable warhead base is projected to be 3000 to 3500 weapons of some seven weapon types. The Department will maintain the capability to reconstitute or reduce this force further. The U.S. has halted the development of new nuclear weapons and has begun closing portions of the weapon production complex and consolidating the remaining elements. In addition, the nation is observing a moratorium on nuclear testing and is pursuing a comprehensive test ban.

However, nuclear dangers remain, and the continued maintenance of a safe and reliable U.S. nuclear deterent is a cornerstone of U.S. national security policy. Thus, the Department of Energy's responsibilities for ensuring the safety, security, and reliability of the U.S. nuclear weapon stockpile will also continue for the foreseeable future.

Meeting these stockpile stewardship and management responsibilities will be more challenging now than ever before, given the moratorium on nuclear testing, the termination of new weapons development, and the closure of production facilities. A new approach to ensuring confidence in the U.S. stockpile is needed. This new approach must rely on scientific understanding and expert judgment, not on nuclear testing and the development of new weapons, to predict, identify, and correct problems affecting the safety and reliability of the stockpile.

Meeting this challenge will be neither inexpensive nor without risk. However, if the strategies laid out by the Stockpile Stewardship and Management Program are followed, both in the near term and the long term, we have confidence that we will be able to successfully meet this new challenge.

In this report, we outline the Department of Energy's Stockpile Stewardship and Management Program. We summarize the technical issues that must be addressed and the enhanced capabilities and facilities that are needed. The program strategy includes the preparation of a Programmatic Environmental Impact Statement under the National Environmental Policy Act to address the environmental impacts associated with the proposed program and a range of reasonable alternatives. In the judgement of the Department, the Stockpile Stewardship and Management Program is essential if the nation is to properly safeguard its nuclear weapons and maintain an unquestioned nuclear deterrent.


In announcing the extension of the nuclear testing moratorium (July 1993), President Clinton reaffirmed the importance of maintaining confidence in the enduring U.S. nuclear stockpile and the imperative to assure that the nation's nuclear deterrent remains unquestioned during a test ban. He clearly acknowledged the need to explore other ways of maintaining confidence in the safety, reliability, and performance of U.S. nuclear weapons. By Presidential Decision Directive and act of Congress (P.L. 103-160), the Department of Energy was directed to "establish a stewardship program to ensure the preservation of the core intellectual and technical competencies of the U.S. in nuclear weapons."

We have developed the Stockpile Stewardship and Management Program to meet the challenges involved in ensuring the safety, reliability, and performance of the enduring stockpile. This program charts a course that must be stayed over the long term to provide responsible and effective stewardship and management of the nation's nuclear deterrent.

Three particular challenges must be met:
* Fully supporting, at all times, the U.S. nuclear deterrent with safe, secure, reliable nuclear weapons while transforming the nuclear weapon complex (laboratories and production facilities) to one that is more appropriate for the smaller enduring stockpile.
* Preserving the core intellectual and technical competencies of the weapons laboratories. Without nuclear testing, confidence in the U.S. nuclear deterrent will rest with confidence in the competency of the people who must make the scientific and technical judgments related to the safety and reliability of U.S. nuclear weapons. A "science-based" stockpile stewardship and management program will enable those people responsible for maintaining the U.S. nuclear stockpile to increase their fundamental understanding of the basic scientific phenomena associated with nuclear weapons.
* Ensuring that the activities needed to maintain the nation's nuclear deterrent are coordinated and compatible with the nation's arms-control and nonproliferation objectives.

Our preliminary cost analysis suggests that, in the absence of a series of ongoing and planned program and management improvements in the way the Department of Energy operates, the Stockpile Stewardship and Management Program would require increased funding after fiscal year 1996. The Department's National Security Five-Year Budget Plan, based on the assumptions used in preparing the fiscal year 1996 budget, projected that without reinvention, funding requirements for the Stockpile Stewardship and Management Program would rise from $3.6 billion in fiscal year 1996 to about $4 billion by fiscal year 1998. The Department is, however, aggressively changing the way it does business. It is anticipated that the Department''s initiatives will lower the funding required during the 1997-2000 fiscal years while accomplishing the Department's national security mission.

The Stockpile Stewardship and Management Program is a single, highly integrated technical program for maintaining the safety and reliability of the U.S. nuclear stockpile in an era without nuclear testing and without new weapons development and production. Traditionally, the activities of the three weapons laboratories and the nuclear test site in Nevada have been regarded separately, and funded separately, from the activities of the weapon production complex. However, all stockpile stewardship and management activities are closely linked to each other, and all are essential to ensure continued confidence in the nation's nuclear deterrent.

Recently, the Department received recommendations from the Galvin Task Force, which provided a comprehensive review of the Department of Energy laboratory system. The Department is committed to carrying out the majority of the Task Force's recommendations and expects that the implementation of these recommendations will result in a laboratory system that is more efficient, cost-effective, and mission-focused. On the question of the Lawrence Livermore National Laboratory, the Department is carefully reviewing the recommended phase-down of Livermore's work in nuclear weapon design and engineering. The timing and details of such a phase-down must depend wholly on how we can best meet our continuing national security responsibilities, as discussed in this report.

Past Approach to Maintaining Stockpile Confidence

The U.S. nuclear weapon stockpile is currently judged to be safe, secure, and reliable. However, decades of experience with the stockpile has often revealed the need for repair or replacement of components and subsystems. Of the weapon systems introduced into the stockpile since 1970, nearly half have required post-development nuclear testing to verify, resolve, or fix problems relating to safety or reliability. Of the seven weapon systems that are candidates for the enduring (START II) stockpile, all seven have already been retrofitted to some degree, including the replacement of major nuclear components in some cases.

The average age of the stockpile has never significantly exceeded the current average age of 12 to 13 years. Although we cannot predict with certainty when age-related changes affecting weapon safety or reliability will occur, we must anticipate they will arise more frequently as the weapons retained in the enduring stockpile age to and beyond their original 20- to 25-year design lifetimes (Figure 1).

Figure 1. Projection of the size and average age of the U.S. nuclear stockpile to the year 2010, assuming current dismantlement plans and current thinking on the composition of the post-START II stockpile. These projections also assume no introduction of new production units and do not account for weapon lifetime extensions.

In the past, a large, often renewed, and diverse stockpile provided "insurance" against single-point and common-mode failures (i.e., failures or defects compromising the safety or reliability of, respectively, a single weapon system or several systems sharing a common design feature). Nuclear testing could be done to provide unambiguous verification of the effects of design features, material changes, or safety issues that could not be adequately calculated or tested in other ways. Continuous development and production of new weapon systems not only provided the U.S. stockpile with the most modern and effective weapons but also maintained the technical competence of the laboratory and production complex in the science and engineering of nuclear weapons. In addition, a steady supply of tritium was provided to support new weapons and to replenish the inventory reduction caused by radioactive decay of the tritium in existing weapons.

Today, none of these conditions exist. Thus it is essential that we develop new strategies and approaches to ensure the safety, reliability, and performance of the stockpile (and confidence in our ability to do so) under current conditions--namely, no nuclear testing and no new weapons in development or production. In addition, we must provide a source of tritium.

Program Goals

The primary goal of the Stockpile Stewardship and Management Program is to provide:
* High confidence in the safety, security, and reliability of the U.S. stockpile to ensure the effectiveness of the U.S. nuclear deterrent while simultaneously supporting U.S. arms-control and nonproliferation policy.

Because the stockpile must endure, the program must provide:
* A small, affordable, and effective production complex to provide component and weapon replacements when needed, including limited-lifetime components and tritium.

Because the world is uncertain and global nuclear threats persist, the program must provide:
* The ability to reconstitute U.S. nuclear testing and weapon production capacities (consistent with Presidential directives and the Nuclear Posture Review), should national security so demand in the future.

Nuclear Test Readiness

The capability to resume underground nuclear testing will be preserved in accordance with the direction of the President. We will document past test data, iterview laboratory people with nuclear testing experience, and archive this information for the future. With a steady program of sophisticated related nonnuclear experiments, we expect to be able to retain and refresh the knowledge and skills needed for nuclear testing.

Critical Scientific and Technical Issues

In developing the Stockpile Stewardship and Management Program, we assessed the scientific and technical issues required to ensure the safety and reliability of the enduring U.S. stockpile. The enduring stockpile was assumed to be the START II stockpile projected by the Nuclear Posture Review. We "stepped forward" to the needs of the stockpile in 2010, when almost all of the candidate weapon systems for the START II stockpile will have reached or exceeded their original design life. Our goal was to identify strategies that would enable us to maintain these weapons continuously, preserve the program-essential technical competence of the weapons laboratories, and transform the existing laboratory and production complex into a smaller, more efficient, less costly complex appropriate for supporting the smaller, less diverse stockpile of the future. The overall program strategy was driven by Department of Defense requirements of the Department of Energy (as identified by the Nuclear Posture Review) to:
* Maintain U.S. nuclear weapons capabilities without nuclear testing or the production of fissile materials and without the production of new-design warheads.
* Ensure the availability of tritium.

Five critical issues were identified, and strategies were developed to address them.

Maintaining Confidence in Stockpile Safety and Reliability without Nuclear Testing

* Nuclear testing provided data sufficient to assess and maintain confidence in the safety and performance of the stockpile weapons.
* Upgraded or new experimental and computational capabilities are needed to fill in those areas of nuclear weapon science that are incomplete, especially gaps in our physics understanding and holes in the data needed for computational simulations of weapon performance and assessments of weapon safety and reliability.
* An improved science-based program with enhanced experimental and computational capabilities is necessary to prevent loss of confidence in the stockpile. This program must be technically challenging so that it will attract the high-quality scientific and technical talent needed for future stewardship of the stockpile.
* The strategy to address this critical issue is discussed under "Enhanced Experimental and Computational Capabilities."

Reducing the Vulnerability of the Smaller Stockpile to Single-Point and Common-Mode Failures

* A large stockpile, with over 20,000 weapons and more than 25 weapon systems, provided substantial protection against single-point and common-mode failures.
* A smaller stockpile, with fewer than 5000 weapons and 7 weapon systems, will be far more vulnerable to single-point and common-mode failures.
* Enhanced weapon and materials surveillance capabilities are necessary to detect potential problems earlier and lessen the vulnerability of the enduring stockpile to these failures.
* The strategy to address this critical issue is discussed under "Enhanced Weapon and Materials Surveillance Technologies."

Providing an Effective and Efficient Production Complex for the Smaller Stockpile

* In the past, a large weapon production complex of seven plant sites provided the capability and capacity to rapidly fix problems in the stockpile.
* Currently, only four plant sites are available (Kansas City, Pantex, Savannah River, and Oak Ridge). Nonnuclear functions are being consolidated and reestablished, and some nuclear functions are not currently available and will be difficult to reestablish. The existing production complex would be inefficient and ineffective for the smaller enduring stockpile.
* Advanced manufacturing and materials technologies must be developed to provide timely and flexible response in correcting stockpile problems. Research, development, and manufacturing must be highly integrated. The capacity-based production infrastructure of the past, which relied heavily on a production complex geared to continuous upgrading and renewal of a relatively large stockpile, must be replaced by a much smaller and more efficient capability-based complex supported by improved scientific understanding of nuclear weapons and their production processes. • The strategy to address this critical issue is discussed under "Effective and Efficient Production Complex."

Providing for Long-Range Support of the Enduring Stockpile

* In the past, continuous development and production of new weapons maintained the scientific and technical knowledge and skills base essential for maintaining the safety and reliability of the stockpile.
* With no new weapons in development or production, budget reductions, and an aging staff with actual experience in designing, testing, and producing nuclear weapons, the knowledge and skills base unique to nuclear weapons will atrophy.
* A new, long-range planning strategy needs to be developed in conjunction with the Department of Defense. This strategy must allow for a weapons complex (design, development, and production) to maintain the U.S. nuclear stockpile and support the nation's nuclear deterrent in the future while meeting our obligation to maintain the safety and reliability of the stockpile while transitioning to a complex more appropriately sized and structured to ensure efficiency and effectiveness in supporting the nation's nuclear deterrent in the future. This long-range strategy should protect the national security option to develop new nuclear weapons.
* The strategy to address this critical issue is discussed under "Long-Range Stockpile Support."

Ensuring an Adequate Supply of Tritium

* All of the candidate weapons for the START II stockpile require tritium replenishment.
* No production source of new tritium currently exists.
* Although the projected START II stockpile and the mandated five-year reserve can be maintained until about 2011 by recycling existing tritium supplies, it will likely take 10 to 15 years to bring a new tritium production source on line. Reactor and accelerator technologies for producing tritium are currently being evaluated, and a decision as to the preferred approach is expected later this year.
* The Department is also developing a contingency option for the production of tritium in the event of a national emergency.
* The strategy to address this critical issue is discussed under "Tritium Production."

Program Strategies

The following strategies have been defined to address the critical issues related to ensuring the safety, reliability, and performance of the enduring U.S. nuclear stockpile.

Enhanced Experimental and Computational Capabilities

Substantial advances in experimental and computational capabilities are needed to maintain confidence in the safety and performance of the U.S. stockpile without nuclear testing. Without nuclear testing as the final "arbiter," we must fill in those areas of nuclear weapon science that are incomplete, particularly gaps in our physics understanding and holes in the data needed for computational simulations of weapon performance and model-based assessments of safety and reliability issues. Upgraded or new experimental capabilities are needed to meet these requirements and, at the same time, to validate improved or new computational models.

Confidence without Nuclear Testing

* Underground nuclear testing provided data sufficient to assess and maintain confidence in the safety and performance of the U.S. stockpile.

* Enhanced predictive capabilities are needed to assess complex problems affecting an aging stockpile.

* Our fundamental understanding of the physics of nuclear weapons must be improved.

Enhanced capabilities will provide the ability to evaluate some safety and performance issues that could have significant stockpile consequences. (It is possible that without enhanced capabilities, some nuclear components exhibiting changes in composition or structure might have to be retired because we would not be able to certify the acceptability of repaired or modified components.) Furthermore, enhanced experimental and computational capabilities, will enable us to maintain the knowledge and skill base that is essential for training new weapons program personnel. The Appendix (p. 14) provides a description of the proposed new facilities and new facilities under consideration to support the Department of Energy's science-based Stockpile Stewardship and Management Program.

Enhanced Computational Capabilities

* Accelerated Strategic Computing Initiative (ASCI)

-- Increases of more that 1000-fold in computational speed and data storage.

-- Full-system, high-fidelity predictive capability.

Enhanced Experimental Capabilities

* Dynamic radiography:
-- Multiple views at multiple times.
-- Higher resolution.

* Lasers and pulsed power:
-- Higher energy and power density.
-- Broad range of pulse widths.

Aboveground Experimental Capabilities
Several aboveground experimental capabilities have been identified that can compensate, at least partially, for the absence of nuclear testing in assessing the performance and safety of nuclear assemblies.

Primaries. Technical issues of particular concern regarding the primary stage of a nuclear weapon include nuclear criticality of the assembly, ignition of the deuterium-tritium boost gas, the three-dimensional shape of the late-time boost-gas cavity, and the effect of the mix of materials into the boost gas on burn and ignition. High-resolution, multiple-time, multiple-view hydrodynamic experiments using simulant materials will be used to define the implosion characteristics and assess primary safety, reliability, and performance. Well-diagnosed pulsed-power and laser-based experiments will also be used to gain an improved understanding of implosion and ignition physics. Pulsed-power and laser facilities could also be used for studies of the effects of materials behavior, including age-related material changes, on primary performance. The data gathered in such experiments will be essential for evaluating new and evolving computational models of primary-stage behavior.

Secondaries. Critical issues for weapon secondary stages relate to the effects on system performance of manufacturing imperfections and age-related changes in materials characteristics. In order to assess the effects of these manufacturing and materials features, we require improved predictive capabilities regarding radiation transport, secondary hydrodynamics, and fusion burn. More complete and more accurate experimental data are needed to improve our understanding of weapon physics issues related to secondary safety and performance. However, the conditions relevant to secondary performance are extremely difficult to create in a laboratory setting, and most data must be extrapolated to the weapons regime, which requires the expert judgment of weapons scientists. Some of the facilities needed to address issues related to primary stages could also be used to investigate secondary-stage issues, but other facilities specific to secondaries are also needed.

Computational Capabilities
To ensure the safety and reliability of the enduring stockpile, computational simulation must fill the void, to the extent possible, left by the termination of nuclear testing. Computational capabilities underpin every aspect of nuclear weapon design, engineering, and evaluation. Because many aspects of nuclear weapons are extremely complex, a fine balance has existed between physical experiments and numerical simulation. Without nuclear testing, numerical simulations will be the principal way of evaluating the safety of nuclear assemblies and the only way of estimating full system performance. Numerical simulations will also provide an essential tie to the data from past nuclear tests, which is an essential element of ensuring the safety and performance of the enduring stockpile without underground nuclear testing. Computational simulation will be an essential (and sometimes the only) means of predicting the effects of materials aging on component and weapon performance.

Increases of more than a thousand-fold in computational speed and data storage are needed to handle simulations of weapon performance and assessments of weapon safety of the required complexity and detail. Improvements are needed in the spatial resolution (fineness of detail) of the simulation and in the number of dimensions that can be modeled (three dimensions, not just one or two). We must also increase the completeness of our physics models, incorporating improved and extended physics data and more complete physics understanding. This effort is closely linked to experimental efforts to provide improved and expanded physics data and to test the predictions of new or evolving computer models.

The goal of the recently established Accelerated Strategic Computing Initiative (ASCI) is to meet these computational requirements. The weapons laboratories will collaborate with industry and other government agencies to reduce the cost of developing these enhanced capabilities. Clearly, the advances required in computational capability and many of the experimental initiatives required for stockpile stewardship and management are interdependent.

Vulnerability of a Smaller, Less-Diverse Stockpile

* More than 20,000 to fewer than 5000 weapons.

* More than 25 to fewer than 7 weapon types.

* All weapons will age beyond their original design life.

Enhanced Weapon and Materials Surveillance Technologies
Because it will be smaller and less diverse, the enduring U.S. stockpile will be more vulnerable to single-point and common-mode failures (i.e., failures or defects compromising the safety or reliability of, respectively, a single weapon system or several systems sharing a common design feature). New surveillance technologies, coupled with enhanced predictive capabilities as to the effects of materials aging on component and weapon performance, are needed so that we can detect potential safety or reliability problems long before they become serious. New technologies and devices are needed for assessing the "health" of stockpile weapons. Without such improvements in our surveillance capabilities, the possibility of a common-mode failure nullifying a large fraction of the enduring stockpile will become a major concern.

Enhanced Surveillance Capabilities

* Nondestructive technology to examine weapon components.

* Sensors built into stockpile weapons to monitor status.

* Predictive models based on materials science.

Materials Science and Technology
We must develop a better understanding of how aging affects a material's physical characteristics. Innovative experimentation will be coupled with enhanced computational capabilities to create predictive models of the effect of age-related materials changes on system performance. Upgraded materials science laboratories will be needed, especially for plutonium and other nuclear materials (e.g., the Chemistry and Metallurgy Research Laboratory at Los Alamos).

Nondestructive Evaluation
Improved nondestructive evaluation techniques, such as acoustic, x-ray, and neutron tomography, would allow us to reduce the number of nuclear components that must be destructively tested to certify their safety and reliability. Of particular importance is the ability to nondestructively examine nuclear assemblies. Facilities such as the Los Alamos Neutron Scattering Center (LANSCE) are able to use high-energy neutron radiography to image low-density materials inside high-density metals and thus may be suitable for surveillance of stockpile nuclear assemblies.

Integrated sensor packages could provide self-diagnostics of weapon components and subsystems, greatly enhancing the effectiveness of stockpile surveillance. The Stockpile Stewardship and Management Program calls for activities to develop sensors that can identify unique signatures (e.g., chemical) that correlate to materials aging concerns. Materials science and microelectronics development laboratories will be essential for the development of such weapon self-diagnostic capabilities.

Hydro Experiments
Hydrodynamic testing on test units built, when possible, with aged stockpile components (with modified pits using simulant materials) will provide important data on the effects of aging on weapon safety and performance. Data from experiments with these test units will augment the baseline data from hydrodynamic tests of stockpile weapons.

Hydronuclear experiments were conducted during the 1958-1961 test moratorium to address stockpile safety concerns. These experiments combined the normal high-explosive trigger for a nuclear device with a quantity of fissile material much less than that required for a nuclear explosion, as the term is usually understood. The Department of Energy currently has no plans to conduct hydronuclear experiments in the 1995 and 1996 fiscal years.

Effective and Efficient Production Complex

In years past, a large weapon production complex provided the capability and capacity to rapidly fix problems with stockpile weapons. Today, elements of the production complex have already been shut down and manufacturing capabilities are being consolidated at fewer sites, which are being downsized as well. At the end of the 1994 fiscal year, the Rocky Flats, Mound, and Pinellas plants completed their last deliveries of products to the stockpile. Production capabilities formerly located at these sites have been or are being established at other sites. However, it will not be practical or cost effective to meet future manufacturing needs by keeping many of the old processes or facilities on "standby."

Large Production Complex Ineffective for Smaller Stockpile

* Facility infrastructure costs are high, especially those for nuclear facilities.

* Some production capabilities and capacities have already been lost.

* Older manufacturing processes do not meet modern environmental, safety, and health standards.

New Approach to Manufacturing
Clearly, a different approach is needed. Investment will be needed to develop manufacturing processes that are agile, safe, and minimize the production of hazardous waste. Most important, attempting to maintain current production capability and infrastructure is not feasible under today's conditions (i.e., no new weapons development or production, the closure and downsizing of existing production facilities, increasingly stringent regulations regarding hazardous waste). In order to ensure the safety and reliability of the enduring U.S. stockpile, appropriate new manufacturing capabilities must be developed that eliminate the need for large facilities and infrastructure.

Manufacturing and Materials Technology

* Integrate research and development with manufactuing activities.

* Capability-based production capacity.

* Low-cost, rapid, and flexible response to fix stockpile problems.

The Stockpile Stewardship and Management Program calls for concurrent engineering-namely, the integrated development of weapon components with the associated advanced manufacturing and materials processes. This integrated approach to design and manufacturing will provide the capability to respond flexibly and rapidly to potential needs for production. It will also be much more cost effective than the sequential approach used in the past. Further cost savings can be realized by using commercially available products and processes wherever appropriate.

Computer-based models to predict the performance of weapon components and to describe the associated manufacturing processes are a key element of concurrent engineering. Models of the manufacturing processes will make it possible to identify process-control signatures, and the models can be used on the factory floor to implement sensor-based adaptive process control during manufacture. Detailed process descriptions, especially when combined with process-control instrumentation, will make it possible to produce the small quantities of reliable, high-quality, specialty products needed to maintain the enduring U.S. stockpile. The advanced computational capabilities developed through the ASCI initiative will help realize the full potential of model-based weapon design and manufacturing.

The Department of Energy has established the ADAPT (Advanced Design and Production Technologies) initiative to develop the tools needed to realize these objectives.


At first glance, remanufacturing of weapon components to their original specifications appears to be a straightforward, low-cost approach to maintaining the stockpile. One could reason that since the designs for the weapons are available, it should be a matter-of-fact process to simply build identical units to replace aging stockpile weapons. However, upon closer examination, this approach is seen to have many drawbacks. Precise replication would not always be possible because deviations from original specifications would occur as a result of interrupted commercial supply bases for specific materials and products. Changes in commercial technology, such as electronics, are frequent, and previously available parts and devices are discontinued. Materials and manufacturing processes are modified to meet more stringent environment, safety, and health standards. In addition, remanufacturing alone would not permit stockpile improvements to address reliability or safety concerns. Perhaps most important, remanufacturing would not retain the required breadth and depth of nuclear weapon expertise and judgment that will be needed to address future concerns about the safety and reliability of an aging stockpile. Therefore, we conclude that remanufacturing alone is not sufficient to maintain and manage the enduring U.S. stockpile.

System Engineering
System engineering-the myriad tasks involved in turning a nuclear assembly into a deployable weapon-is closely linked with production capabilities and therefore must be integrated with this new approach to manufacturing and materials. In the past, weapon system and subsystem engineering relied on prototype testing, with sequential steps of design, testing, and evaluation, followed by design modification, further testing, and further evaluation. For the future, system engineering will rely more extensively on computer modeling to predict system, subsystem, and component performance before any hardware is produced. Computer-based models will also be used extensively to evaluate the safety and reliability of weapon systems when exposed to a wide range of abnormal environments, such as severe accidents. The number of costly prototypes that must be fabricated and tested will be greatly reduced, making this computer-based approach to system engineering more cost effective. In addition, integrating model-based development of manufacturing processes into the early stages of engineering ensures the producibility of the component or system and makes it possible to use cost as a factor for selecting among alternative designs. Effective implementation of this approach requires increases in computational speed and memory and thus is dependent on the ASCI initiative.

A number of unique test facilities must be retained and selectively enhanced, including environmental test facilities for evaluating model predictions, certifying performance, and assessing safety. Test ranges for conducting joint flight tests with the military to evaluate weapon system performance under realistic weapon delivery conditions will still be required. Enhanced experimental capabilities may also be needed to evaluate weapons and components in combined environments (e.g., heat plus shock) or to certify the hardness of replacement components to neutrons, gamma rays, and x rays.

Advanced Production Technology

* Computer-generated solid models of products.

* Electronic information about material properties.

* Predictive computer models of manufacturing processes.

* Sensor-based adaptive process control of manufacturing.

Long-Range Stockpile Support

As long as the U.S. has a nuclear weapon stockpile, there will be a need to evaluate periodically the safety, security, and reliability of each of the weapon types in the stockpile, to repair them when concerns arise, to upgrade them to meet more demanding safety and security standards and new military requirements, to replace them when refurbishment is not practical or cost effective, and to dismantle them at the end of their useful lifetime. Many of the capabilities needed for long-term support of the stockpile currently exist and will be maintained. However, as a result of the closure of elements of the production complex, some capabilities need to be reestablished and others need to be enhanced.

Long-Range Stockpile Support

* In the past, continuous design, development, and production of nuclear weapons "stabilized" the laboratory and production complex.

* Without a steady flow of new weapon development projects, a different approach to long-range stockpile support is needed.

Sufficient stockpile dismantlement capabilities currently exist. From fiscal year 1991 to the present, the Department of Energy has dismantled more than 7000 nuclear weapons. The capability-specifically the technical expertise and unique facilities currently located at the Pantex Plant and the design expertise and judgment only available at the weapons laboratories-must be sustained to meet future dismantlement requirements.

Routine Stockpile Surveillance
Routine stockpile surveillance activities will continue. Each year, a statistically significant fraction of every weapon type in the active stockpile will be evaluated and tested. We will use laboratory tests, which evaluate conditions and performance of individual components against design requirements, as well as flight tests of modified warheads (in which fissile materials are replaced with simulant materials), which verify nonnuclear performance in a delivery environment. Currently, more than a hundred weapons are removed annually from the stockpile for laboratory and flight tests. For the aging, less-diverse enduring stockpile, new evaluation techniques must be developed to identify and deal with common-mode failures that could negate a significant fraction of the stockpile.

Corrective Maintenance and System Replacement
In the past, safety and reliability concerns could be addressed by replacing older weapon systems with new ones incorporating modern safety, security, and performance features. Since there are no new weapons in development or production, this approach is no longer possible. Instead, we will retrofit, upgrade, or replace stockpile weapons to ensure their continued safety, security, and reliability. Indeed, one of the weapon systems slated for the enduring stockpile is currently being retrofitted at Pantex with safety, security, and reliability upgrades, and several other field and Pantex retrofits are planned for the next five years.

The Stockpile Stewardship and Management Program calls for new approaches to ensuring the ability to fix problems that will undoubtedly occur in the aging stockpile. To start with, safety margins will be increased (which may include modifications to primaries), use-control technology will be enhanced, and components produced with "sunset" technologies (i.e., products and processes that become obsolete because of, for example, increasingly stringent environmental or safety regulations) will be replaced. To accomplish these objectives cost effectively, we will conduct planned product-improvement programs that will significantly advance the safety, security, reliability, and/or maintainability of stockpile weapons. To extend the lifetime of weapon components without jeopardizing safety or reliability, we will use improved predictive capabilities, made possible by a combination of computational modeling and experimentation, to define age-related changes in materials properties and will engage in preventative maintenance (before a problem develops) of the stockpile. This approach has already been used successfully for such limited-lifetime components (LLCs) as neutron generators and tritium reservoirs, and it may be possible to extend this approach to entire weapon systems. Not only will these activities ensure and improve the safety and reliability of the enduring stockpile, but they will also exercise and sustain much of the skill base required for nuclear weapon development and production and thus help maintain the nation's nuclear competency.

Example of Preventative Maintenance--Neutron Generator

* Repair:
--Correct detonator corrosion problem.

* Life extension:
--Review tube life data annually.

--Develop new generator with improved reliability and durability (in progress).

Improvements to the stockpile must be made through a strategy that does not call for significant new weapons production or complete rebuilding of the candidate weapons for the START II stockpile. It is not cost effective, and perhaps not even feasible, to simply replace individual weapons when they reach the end of their original design lifetime (20 to 25 years, depending on weapon type). A new strategy will be developed in conjunction with the Department of Defense. It will rely extensively on preplanned lifetime extension projects and, whenever a major retrofit is not cost effective, on one-for-one complete system replacements. The retrofit/rebuild schedules will be phased to level the production and recertification workload and to ensure sustainable effectiveness of design, engineering, and production capabilities in every critical area.

Tritium Production

Tritium is required for all weapons in the enduring U.S. stockpile. This radioactive isotope of hydrogen has a half-life of 12.5 years and decays at the rate of about 5% per year. Tritium has not been produced in the U.S. since 1988. Stockpile tritium requirements are currently being met by recycling the tritium from dismantled weapons. Recycling will meet the tritium requirements of all of the weapons in the START II stockpile, including a five-year reserve, until about 2011. Clearly, some means of tritium production will be required to support the stockpile after that time.

Various tritium production technologies are being evaluated by the Department of Energy, and the preferred technology will be selected in the near future. The following technologies for producing tritium are being considered: accelerator, advanced light-water reactor, heavy-water reactor, and modular high-temperature gas-cooled reactor. Candidate sites for tritium production and recycling are the Idaho National Engineering Laboratory, the Nevada Test Site, the Oak Ridge site (Tennessee), the Pantex Plant (Texas), and the Savannah River Plant (South Carolina). It is estimated that, regardless of the technology and site chosen, it will take 10 to 15 years to establish a new tritium production capability. The current plan provides for a technically proven contingency supply of tritium, based on a commercial light-water reactor, to respond in the event of a national emergency.

The tritium recycle and supply proposal has its own programmatic environmental impact statement. A record of decision on the technology and siting is scheduled for this year.

Program Costs

Budget projections derived from a preliminary analysis of the Stockpile Stewardship and Management Program are shown in Figure 2. The rise in funding for the next several years is the investment that would be required to implement this science-based Stockpile Stewardship and Management Program in the absence of the Department's reinvention efforts to improve efficiency and reduce costs.

Stockpile stewardship costs include all research and development activities at the weapons laboratories to implement the program strategies and maintain the Presidentially directed test-readiness posture at the Nevada Test Site. The stockpile stewardship cost estimate includes the proposed new experimental test facilities (the Dual Axis Radiographic Hydrodynamic Test Facility, the Contained Firing Facility, the Atlas Facility, the National Ignition Facility, and the Process and Environmental Technology Laboratory; see the Appendix, p. 14, for descriptions of these facilities). Stockpile management costs include a new tritium production facility and the activities associated with dismantling the nonenduring stockpile, with maintenance and surveillance of the enduring stockpile, and with the implementation of new manufacturing and surveillance technologies to support the enduring stockpile.

Figure 2. Budget projection for the Stockpile Stweardship and Management Program in budget-year dollars for the fiscal years 1991 through 2000 before reinvention (costs for 2001 and beyond are in constant fiscal year 1996 dollars). It is anticipated that the Department's initiatives will reduce the funding required for fiscal year 1997 and beyond. Costs for the various elements of the program are cumulative.

These budget projections do not include the following: worker retraining, science education support at public schools, or national economic competitiveness not critical to nuclear weapons. Neither do they include the cost of decommissioning and decontaminating facilities, direct funding for nonproliferation and arms control activities, or the cost of storage and disposition of excess nuclear components and materials after fiscal year 1997.

Our preliminary cost analysis suggests that, in the absence of a series of ongoing and planned program and management improvements in the way the Department of Energy operates, the Stockpile Stewardship and Management Program would require increased funding after fiscal year 1996. The Department's National Security Five-Year Budget Plan, based on the assumptions used in preparing the fiscal year 1996 budget, projected that without reinvention, funding requirements for the Stockpile Stewardship and Management Program would rise from $3.6 billion in the 1996 fiscal year to about $4 billion by fiscal year 1998. The Department is, however, aggressively changing the way it does business. It is anticipated that the Department's initiatives will lower the funding required during the 1997-2000 fiscal years while accomplishing the Department's national security mission.

Programmatic Environmental Impact Statement

The Department of Energy has committed to do a programmatic environmental impact statement (PEIS) for the Stockpile Stewardship and Management Program. To assist in the development of this PEIS, the Department plans to hold an information meeting in the Washington, D.C., area prior to the "notice of intent" and to involve stakeholders through formal scoping meetings in the vicinity of the three weapons laboratories, the four remaining production plant locations, and the Nevada Test Site. A tentative schedule for the PEIS follows.
May 1995....................................................................Information meeting
June 1995....................................................................Notice of intent
June-August 1995........................................................Scoping meetings
September-October 1995............................................Implementation plan
January-February 1996................................................Draft PEIS
July-August 1996.........................................................Final PEIS

Summary: Maintaining Confidence in the Safety and Reliability of the Enduring U.S. Nuclear Weapon Stockpile

We must take positive steps to preserve the current high confidence in the safety and (to the extent possible without nuclear testing) the performance of the enduring stockpile. This raises the questions of how one measures confidence and what level of confidence is desired. Confidence is subjective and, as such, rests on the judgment of people. Judgment is based on information, experience, and trust in the sources of the information and experience. This link between confidence, judgment, and people is the reason that the competency and experience of our weapons scientists and engineers are so crucial to the U.S. nuclear weapons program. As a result, maintaining this competency base, which exists primarily at the weapons laboratories, is one of the highest priorities of the Stockpile Stewardship and Management Program. We need to preserve and pass on the competency base developed during the years when nuclear testing was permitted. It is this need that drives our efforts to retain the staffs of test-experienced weapons scientists and engineers and to attract talented new people to the program. In turn, these efforts are behind our push to maintain the weapons laboratories with their reputations for scientific and technical excellence, to engage in research and development programs that are nationally important and technically challenging, and to support state-of-the-art experimental facilities and technical capabilities.

Peer Review

Throughout the history of the U.S. nuclear weapons program, we have compensated for less-than-complete knowledge about the physical phenomena governing nuclear weapon operation with nuclear testing and interlaboratory peer review. Without nuclear testing, confidence in the safety and reliability of U.S. nuclear weapons must be based more on judgment and less on data. Peer review is more important now than ever before because judgment-based confidence relies on the knowledge that the judgment has withstood scrutiny. Thus it is essential that we preserve an independent review process.

Clearly then, preserving high confidence in the safety and performance of the enduring U.S. stockpile without nuclear testing will require an improved, more complete, more accurate understanding of the underlying physical principles involved in nuclear weapons. In turn, this will require new experimental capabilities and greatly improved computational capabilities. As these new capabilities come "on line" and we gain experience in their use and demonstrate the validity of the information they provide, we believe we will be able to certify the safety and performance of the stockpile. (Figure 3 summarizes the activities of the Stockpile Stewardship and Management Program.)

Science-based stewardship and management of the U.S. stockpile has never been done before. Meeting this challenge will be neither inexpensive nor without risk. However, if the strategies laid out by the Stockpile Stewardship and Management Program are followed, both in the near term and the long term, we have confidence that we will be able to successfully meet this new challenge.

Figure 3. Summary of the facility closure, consolidation, and construction activities of the Stockpile Stewardship and management Program.

Appendix: New Facilities

Proposed New Facilities

The following new facilities are included in the Department of Energy's National Security Five-Year Plan. Construction and operation of these facilities is subject to approval at the appropriate Key Decision stages of the Department's major systems acquisition process and to future budget decisions.

Dual Axis Radiographic Hydrodynamic Test Facility
The Los Alamos Dual Axis Radiographic Hydrodynamic Test (DARHT) facility would provide substantial improvements in dynamic radiography, which is a major experimental tool for addressing issues of weapon safety and reliability and for validating our physics understanding of and predictive capability for primaries. This facility would produce radiographic images with significantly higher spatial resolution and illumination intensity than are possible with present facilities. The dual-axis capability of DARHT would provide data on implosion symmetry as a function of time. Improvements in hydrodynamic testing capabilities, by which we can probe the implosion of primaries and assess other weapon hydrodynamic effects, are key to our ability to address stockpile reliability and safety issues by means other than nuclear testing. These hydrotesting improvements are also essential for validating improved numerical codes and models of weapon performance.

The construction of DARHT has been halted by court action. Whether or not construction is resumed will depend on the outcome of an environmental impact statement now being prepared and a record of decision that will follow.

Contained Firing Facility
The Contained Firing Facility (CFF), an addition to the Flash X-Ray (FXR) hydrodynamic testing facility at Lawrence Livermore, would provide for well-diagnosed, contained hydrodynamic tests with up to 60 kg of energetic explosives as well as new diagnostics for improved studies of the behavior of weapon materials under explosive shock conditions.

National Ignition Facility
The National Ignition Facility (NIF) would simulate, on a small but diagnosable scale, conditions of pressure, temperature, and density close to those that occur during the detonation of a nuclear weapon. With the NIF, it would be possible in the laboratory, for the first time ever, to study radiation physics in a regime close to that of secondaries. The NIF would be used to investigate hydrodynamic and mix phenomena relevant to modern nuclear weapons. The NIF would also provide a unique laboratory capability for studying thermonuclear ignition and burn of dense deuterium-tritium gas. Furthermore, by studying NIF-heated targets, we would be able to improve our ability to predict the effects of x-radiation on weapon components and weapon systems; these studies would be a valuable complement to experiments at other Department of Defense and Department of Energy facilities.

Without the capabilities offered only by the NIF, uncertainties about some physics areas that affect secondary performance would go unanswered, and improvements in our predictive capabilities would suffer. Equally important, without the NIF and similar frontier-expanding facilities, the weapons laboratories would find it increasingly difficult to maintain the necessary expertise and skill bases unique to nuclear weapons and essential for science-based stockpile stewardship and management.

Atlas Facility
The pulsed-power Atlas Facility at Los Alamos would provide implosions of centimeter-scale metallic shells for high-resolution experiments related to secondary hydrodynamics. It would also provide a long-pulse-width, soft-x-ray source for radiation-flow experiments scaled to secondary channel conditions. In addition, Atlas would be used for large-scale hydrodynamic experiments with high-fidelity primary pressures to resolve issues related to implosion stability and boost-gas mixing. Atlas would help us gain a more complete understanding of hydrodynamic and radiation-flow phenomena, improve our predictive capabilities, and assist in our efforts to ensure essential nuclear weapon skill and judgment bases.

Process and Environmental Technology Laboratory
The Process and Environmental Technology Laboratory (PETL) at Sandia would support efforts in advanced and environmentally benign manufacturing, specialty materials formulation, materials aging, and analysis. This materials research facility would also support almost all aspects of nonnuclear component development and production engineering. It would be used for the development of advanced stockpile surveillance technologies, the assessment of age-related defects identified during routine surveillance, and the development of cost-effective manufacturing technologies for retrofit and replacement components. The PETL would replace a number of existing facilities distributed throughout Sandia, some of which would require modification to meet current safety standards in order to be used for planned stockpile stewardship activities. Upgrading existing facilities to meet current safety standards would cost more than the cost of constructing this new facility; even with upgrades, existing facilities would not be able to provide the improvements in operating efficiencies offered by the PETL.

New Facilities under Consideration

A number of additional new facilities are under consideration to meet the challenges of science-based stockpile stewardship and management. Depending on the outcome of the PEIS process, other facilities or major upgrades of existing facilities may be needed.

Integrated Design and Development Center
The Integrated Design and Development (ID&D) Center at Sandia would enable the development and validation of tools for establishing the low-cost supply of nonnuclear components (needed to maintain or upgrade stockpile weapons). It would enable integrated optimization of product designs and manufacturing processes to ensure rapid product realization of highly reliable components at the lowest possible cost. The ID&D Center would also serve as a testbed for rapid prototyping and as a source for specialized products not available from industry.

Advanced Hydrotest Facility
The Advanced Hydrotest Facility (AHF) would provide up to eight radiographic views (compared to the two views that would be provided by DARHT). Such multiple-view, multiple-time radiography is anticipated to be essential for assuring weapon reliability and safety without nuclear testing in the long term. The AHF would provide multiple images (20 or more) that would reveal the evolution of a primary's implosion symmetry and boost-cavity shape under normal conditions and in accident scenarios. The AHF would be based on new and developing accelerator technology, which draws on the rapidly advancing state of the art in high-power, high-speed, solid-state components.

Jupiter Facility
The Jupiter Facility would be an advanced pulsed-power x-ray source to provide enhanced capabilities in areas of weapon physics, radiation-effects science, and pulsed-power technology. It would provide a class of x-ray environments that can be obtained elsewhere only in underground nuclear tests. This facility would enhance our ability to certify that critical weapon components meet military requirements for x-ray hardness.

High-Explosive Pulsed-Power Facility
The High-Explosive Pulsed-Power Facility (HEPPF) would provide experimental capabilities for studying secondary physics issues at shock pressures and velocities approaching those of actual weapon conditions at scale sizes and pulse widths needed to validate numerical simulations. High-explosive pulsed power is a cost-effective means for producing the extraordinary electrical pulses needed for these experiments. This facility would significantly enhance our predictive capabilities regarding secondary aging and remanufacturing. Laboratory Integrated Simulation and Computer Center The Laboratory Integrated Simulation and Computer Center (LISAC) would replace the 1950s facility currently housing Sandia's mainframe computers. It would increase the efficiency and effectiveness of Sandia's computer operations and provide a collaborative setting for the development of advanced software, operating systems, and peripheral equipment. The highly interactive environment made possible by the LISAC would also help attract computer experts to Sandia for extended temporary assignments, a central element of the Accelerated Strategic Computing Initiative (ASCI).


agile manufacturing
Effective use of equipment and people, involving time-sharing of expensive equipment and broader responsibilities for workers on the manufacturing floor, to rapidly and cost effectively meet the needs of high-quality, low-volume production.

The process by which fusion of deuterium-tritium gas inside the pit of a nuclear weapon produces neutrons that increase the fission output of the primary.

Fusion of two light nuclei (usually deuterium and tritium) to form a heavier nucleus (helium) accompanied by the release of neutrons and energy.

common mode failure
A failure or defect affecting an entire class of weapon or weapon component; a particular concern with the enduring stockpile since it contains about seven weapon systems, many of which use identical components, components with common design features, or components manufactured using identical or similar processes.

computational modeling
The use of a computer to develop a mathematical model of a complex system or process and to provide conditions for testing it.

concurrent engineering
Concurrent design of both the product and the processes for manufacturing the product; integrated design, production, prototyping, and product qualification.

An exothermic chemical reaction that propagates with such rapidity that the rate of advance of the reaction zone into the unreacted material exceeds the velocity of sound in that material; that is, the advancing reaction zone is preceded by a shock wave.

Something that is based on actual measurement, observation, or experience rather than on theory.

enduring stockpile
The U.S.nuclear stockpile of the future, consisting of about seven weapon systems (many of them older than their design lifetime), with no new systems added to the stockpile for the foreseeable future.

explosion (conventional)
A chemical reaction or change of state that occurs in an exceedingly short time with the generation of high temperatures and large quantities of gaseous reaction products.

explosion (nuclear)
An explosion for which the energy is produced by a nuclear transformation,either fission or fusion. The term typically implies the release of enormous amounts (kilotons) of energy.

Nuclear reaction in which a heavy nucleus is split apart to form two or more lighter nuclei, accompanied by the release of large amounts of energy and various forms of ionizing radiation.

Nuclear reaction in which light nuclei are fused together to form a heavier nucleus, accompanied by the release of immense amounts of energy and fast neutrons.

high explosive
An energetic material that detonates (instead of deflagrating or burning); the rate of advance of the reaction zone into the unreacted material exceeds the velocity of sound in the unreacted material.

hydrodynamic test
High-explosive nonnuclear experiment to investigate hydrodynamic aspects of primary function up to mid to late stages of pit implosion.

The study of the motion of a fluid and of the interactions of the fluid with its boundaries, especially in the case of an incompressible inviscid fluid.

hydronuclear experiment
Very-low-yield experiment (less than a few pounds of nuclear energy released) to assess primary performance and safety with normal detonation.

Self-sustained fusion burn of light nuclei.

The sudden inward compression and reduction in volume of fissionable material with ordinary explosives in a nuclear weapon.

limited-lifetime component
A weapon component that decays with age and must be periodically replaced.

Integrated circuits and electronic devices constructed of individual circuit elements with dimensions of micrometers (10-6 m) on a carrier with dimensions of a centimeter (10-2 m).

Mixing of materials, usually with different densities and velocities, that can adversely affect nuclear weapon performance.

nondestructive evaluation
Test method that does not involve damage to or destruction of the test sample; includes the use of ultrasonics, radiography, magnetic flux, and other techniques.

noninvasive imaging
Imaging method that does not damage the test specimen; includes radiography, computed tomography, and other techniques.

nonnuclear component
Any one of thousands of parts that do not contain radioactive or fissile material that are required in a nuclear weapon.

Preventing the spread of nuclear weapons, nuclear weapon materials, and nuclear weapon technology.

nuclear assembly
Collective term for the primary, secondary, and radiation case.

nuclear component
A part of a nuclear weapon that contains fissionable or fusionable material.

nuclear weapon
The general name given to any weapon in which the explosion results from the energy released by reactions involving atomic nuclei, either fission, fusion, or both.

nuclear weapons complex
The collection of laboratories and production plants involved in the design, production, and testing of nuclear weapons.

nuclear warhead
A warhead that contains fissionable and fusionable material the nuclear assembly and nonnuclear components packaged as a deliverable weapon.

numerical simulation
The use of mathematical algorithms and models of physical processes to calculationally simulate the behavior or performance of a device or complex system.

The ability of a nuclear weapon or weapon system to operate in specified manner (e.g., yield, range, accuracy, radiation spectrum) under stated conditions. (Essentially equivalent to reliability.)

The ability of a nuclear weapon, weapon system, or weapon component to perform its required function under stated conditions for a specified period of time. (Essentially equivalent to performance.)

To furnish (e.g., a weapon) with new parts, equipment, or features not available at the time of manufacture.

Minimizing the possibility that a nuclear weapon will be exposed to accidents and preventing the possibility of nuclear yield or plutonium dispersal should there be an accident involving a nuclear weapon.

Minimizing the likelihood of unauthorized access to or loss of custody of a nuclear weapon or weapon system, and ensuring that the weapon can be recovered should unauthorized access of loss of custody occur.

simulant material
Materials used to modify a weapon pit to prevent the device from becoming critical.

special nuclear material
A specific list of materials including plutonium and uranium, among others.

stockpile management
The specific tasks and functions involved in managing the stockpile weapons, including production, routine surveillance and servicing, assembly and dismantlement, and disposal of weapons-related parts and materials.

stockpile stewardship
The science and technology aspects of ensuring the safety, security, and reliability of the stockpile, including research and development to provide the technologies required for stockpile management.

stockpile surveillance
Routine and periodic examination, evaluation, and testing of stockpile weapons and weapon components to ensure that they conform to performance specifications and to identify and evaluate the effect of unexpected or age-related changes.

Umbrella term for safety, security, and use control.

system integration
The process by which individual components are engineered into a system that meets performance requirements.

test readiness
Maintaining the critical technologies, staff skills, and infrastructure to be able to resume nuclear testing if and when mandated by the President.

The process by which very high temperatures are used to bring about the fusion of light nuclei, such as deuterium and tritium, with the accompanying release of energy.

use control
Delaying or preventing the unauthorized use of a nuclear weapon or weapon system while allowing timely authorized use.

Collective term for the package of nuclear assembly and nonnuclear components that can be mated with a delivery vehicle or carrier to produce a deliverable nuclear weapon.

weapons laboratories
Colloquial term for the three Department of Energy national laboratories-Los Alamos, Lawrence Livermore, and Sandia-that are responsible for the design, development, and stewardship of U.S. nuclear weapons.

weapon system
Collective term for the nuclear assembly and nonnuclear components, subsystems, and systems that comprise a nuclear weapon.