Nuclear Forensics: A Capability at Risk (Abbreviated Version) (2010)

When the U.S. government interdicts significant special nuclear or radioactive materials, the President of the United States wants answers to several questions: What is it? Where did it come from and whose is it? Who had it and how did they get it? Did they have help? What were they going to do with it? Is there more of it out there? What should we do about it?

If the United States were attacked with a nuclear or radiological device, the political environment would be extraordinarily intense, and the pressure for the nation to respond would be great. The president would seek answers to a similar set of immediate and urgent questions, including: What was it? How bad is the damage, and how much worse will it get? Who did it? Are there more out there? Did they have help? Where did it come from? Was it ours? And ultimately, What should we do about it?

The United States has technical capabilities to analyze interdicted nuclear and radiological devices and the materials that can be used to make them, analyze the signals and debris from a nuclear detonation or radiological dispersion, and simulate materials production and weapons performance. Together, these nuclear forensics capabilities can help answer the president's questions by learning what the materials are; how, when, and possibly where they were made; what has been done with them since they were made; and, in the case of a nuclear device, features of the device's design, construction, and performance. Each and any piece of information that nuclear forensics provides could be useful in the investigations of the incident as law enforcement and intelligence agencies work toward prevention, mitigation, and attribution. These capabilities can be made available through the Department of State to other countries, if another nation requests such assistance, as is currently done in conventional terrorist events, such as embassy bombings.

Substantial nuclear forensics capabilities exist in U.S. laboratories today. This report, requested by the Department of Homeland Security (DHS), the National Nuclear Security Administration (NNSA), and the Department of Defense (DoD), makes recommendations on how to sustain and improve those capabilities.

During the days of U.S. nuclear weapons testing, the national security laboratories analyzed measurements taken during and after each test, which provided insights into the phenomena that occurred during the detonation. Some measurements were analyzed quickly, but full results were provided over the course of months. The analyses of these tests led to an understanding of nuclear weapons design and performance that now enables the weapons

laboratories and U.S. Strategic Command to certify the safety and reliability of the U.S. nuclear weapons stockpile without additional nuclear testing. Although this experience is valuable for nuclear forensics, it is much more challenging to work from observations of unknown materials or from detonation debris from unknown devices and analyze and interpret them accurately and quickly. The nuclear forensics task is subject to the challenge of intense time pressure because of the demand for immediate answers. Additionally, in a domestic post-detonation event, the pressure for release of preliminary or partial analyses would be enormous.

In real-world interdiction cases and in exercises, U.S. laboratories have demonstrated the ability to analyze uranium and plutonium samples and report on the composition and time since last chemical separation of the material. In the same samples they can identify other materials and characteristic features of processes and features of facilities that might have produced the material in the sample. The laboratories have extensive experience with many other radioactive materials aside from uranium and plutonium, but most of that experience is not in conducting analyses needed for nuclear forensics, such as trace-constituent analyses and time since last separation. The laboratories are taking steps to improve their practices and methods to build greater confidence in the results. The timelines for obtaining results also vary depending on the material, the type of analysis, and the resources devoted to analysis and evaluation.

In addition to the incidents and domestic exercises, the national laboratories also take part in international round robin exercises that provide opportunities to compare analytic procedures and results. These international efforts complement the U.S. exercises and build relationships among the analysts, enhance development and application of techniques.

In post-detonation exercises, nuclear forensics has provided information sequentially about the nuclear explosive, beginning almost immediately and working to completion after receipt of samples at the national laboratories. For these exercises, the national laboratories were provided with advance notice and relatively good input data.

The timeline for post-detonation analysis and evaluation is longer than is desired. Even delivering on the exercised pace will be taxing to the limited human and facility resources that are available for this operation. It is the committee's judgment that under similar conditions the time required to provide all of the information with higher confidence than can be achieved today could be reduced with improvements in procedures and techniques.

These improvements include:

• development and deployment of prompt diagnostic systems;

• simulations to better interpret data from prompt diagnostics;

• identification of useful short-lived signatures¹ and how to measure them;

• better planning for air and ground sampling;

• development of tools and procedures for sample selection;

• formalizing procedures for faster sample distribution to and receipt by fixed laboratories;

• development of faster, more reliable sample-preparation techniques;

• development of automated analytical techniques that meet modern environmental, health, and safety requirements;

• simulations to explore signatures of nuclear-material-production technologies; and

• more detailed simulations of the performance of a broad range of known and

improvised nuclear devices in the types of environments that are the most likely targets of a nuclear attack.

Although U.S. nuclear forensics capabilities are substantial and can be improved, right now they are fragile, under resourced, and, in some respects, deteriorating. Without strong leadership, careful planning, and additional resources, these capabilities will decline.

Major areas of concern include the complex organization of nuclear forensics efforts within the federal government; the sustainability of a nuclear forensics capability that relies on base support from a shrinking nuclear weapons program for facilities, equipment, and personnel; a diminishing workforce and aging infrastructure; and reliance on procedures and tools developed during the Cold War, which may not be optimized for the nuclear forensics mission. Each of these is described in more detail below along with the committee's major findings and recommendations, which appear in bold text.

The nuclear forensics mission is spread among multiple offices in several federal agencies and is overseen by and reports to different committees and subcommittees in Congress. The Executive Office of the President has assigned leadership for coordinating nuclear forensics to DHS. The National Technical Nuclear Forensics Center (NTNFC) in the Domestic Nuclear Detection Office of DHS was established in October 2006. The NTNFC was given the mission to coordinate the nation's nuclear forensics program. Under the direction of the Secretary of Homeland Security, NTNFC is to develop and implement a nuclear forensics program plan; ensure national-level integration of the program with relevant work performed for intelligence and arms control purposes; assess the capabilities through exercises; build support for improvement in infrastructure, personnel, research, development, testing, and evaluation; and establish standards and stewardship for the nation's nuclear forensic capabilities.

DHS and the other cooperating agencies have not yet devised or institutionalized a program that is optimized with respect to readiness, operational effectiveness, sustainment, and improvement. There is extensive and effective information sharing among several of the parties that constitute the NTNFC. Part of this success results from the sense of common purpose shared by the principals involved, and some arises from the short-term exchanges and extensive use of detailees among agencies. However, DHS does not direct resources or actions in program elements outside of the pre-detonation nuclear materials mission area, and agencies are more likely to optimize within their mission areas than across the mission areas. With this structure, it is difficult to form a coherent program with a single set of common goals and funds aligned appropriately and consistently to meet those goals. In particular, it is difficult to prioritize among missions, actions, and efforts (e.g., research, training, and operations; pre-detonation versus post-detonation capabilities; improvised nuclear devices versus radiological dispersal devices).

The program managers also do not have clear direction from senior decision makers, law enforcement, and the intelligence community on the operational requirements for nuclear forensics and information deliverables. Technical practitioners learn only anecdotally what senior decision makers expect and want most from nuclear forensics. A clearer statement of national-level requirements would provide a basis for interagency coordination and prioritization. To put together a program plan, DHS must work with senior decision makers, law enforcement experts, and members of the intelligence community, as well as the nuclear

forensics practitioners, to define more clearly and build greater acceptance of the program's requirements and goals. This might be begun by convening a workshop composed of experts within these communities with the specific objective of refining and agreeing on a proposed set of requirements and goals to guide the program. Such a workshop could draw upon this and other studies of nuclear forensics functions in the United States, and lessons learned from exercises and real-world cases. Such a workshop is a starting point, but it might not resolve actual disagreements between and among agencies. For such situations, DHS and possibly the Executive Office of the President will need to adjudicate disagreements.

Clarifying the goals for the program is an essential step in developing and refining objectives and guidelines from which to build the capability and performance metrics. Clarifying requirements and other guidance would also inform and support a more realistic assessment of needs and acquisition of resources for near-, mid-, and far-term investments. At each site the committee visited, there were important questions about program needs that the scientists and managers could not answer specifically because they did not have a requirements document. The committee has been told that the last draft requirements document failed to achieve concurrence from two agencies, and that there have been renewed efforts to address the concerns raised and put something in place to assist the program. It makes sense for this to be done in time for an implementation plan that flows from the requirements to affect the next budget cycle.

The organizational complexity of nuclear forensics responsibilities and authority in the U.S. government hampers the program and could prove to be a major hindrance operationally. To address this problem, the committee considered recommending consolidating authority and responsibility for nuclear forensics in a particular agency. Ultimately, the committee did not do so because (l) the committee was not asked to tell the government how to reorganize itself for this mission; (2) the committee could not evaluate the implications, advantages, and disadvantages of doing so; and (3) a presidential directive was issued in 2007 establishing the agencies' responsibilities. However, in the committee's view, the organizational complexity remains a serious problem that must be addressed.

Recommendation 1: If the nuclear forensics mission is to succeed, then: (a) the organizational structure must be streamlined with the agency in charge possessing both the requisite authority and responsibility; and (b) DHS, the cooperating agencies, and the President's National Security Staff must develop and issue the appropriate requirements documents.

The existing nuclear forensics capabilities are highly leveraged off of the nuclear weapons program and other related programs. The system of laboratories, equipment, and personnel upon which they depend was developed and funded by the nuclear weapons program and others. The weapons program's funds have been declining and NNSA is focusing remaining resources more narrowly to eliminate redundancies and maintain stockpile-mission-critical competencies at fewer sites. In these circumstances, the nuclear weapons program may not provide the resources necessary to sustain the nuclear forensics mission, and as a result the laboratory system's effectiveness and responsiveness for nuclear forensics missions become less certain. Historically, most nuclear-forensics functions have been paid for at the marginal cost,

relying on other programs to establish the infrastructure and facilities and to pay for most of the training and the balance of the salary for the workforce, as well as some of the equipment. Because of that reliance, even level funding for nuclear forensics will result in a decline in the nuclear forensics capabilities as the underlying nuclear weapons program shrinks.

Recommendation 2: DHS and the cooperating agencies should issue an implementation plan for fulfilling the requirements and sustaining and improving the nuclear forensics program's capabilities. This plan would represent a coordinated, integrated program view, including prioritized needs for operations, infrastructure, research and development. The plan should specify what entity is responsible for each action or program element. The plan would form the basis for the multi-year budget requests essential to support the program and its plan.

If the implementation plan is reasonable and Congress agrees that nuclear forensics is an important element of our national security, then Congress should fund and support this mission.

To maintain appropriate facilities and equipment, the nuclear forensics program needs:

• Multiple programmatic uses of the facilities and equipment relevant to nuclear forensics so that they are functioning, calibrated, and ready when they are needed. This would result in facilities and equipment whose costs are shared.

• A quality management program, including standards, validated techniques, and regular calibration.

• An on-going effort to modernize and accelerate methods for analysis and evaluation based on R&D.

At present, personnel skilled in nuclear forensics at the national laboratories are too few and are spread too thinly. Furthermore, a substantial fraction of the experienced personnel are retired, now eligible for retirement, or nearing retirement age. The university pipeline produces too few people in needed specialties and universities will not produce them without stable funding for relevant R&D.

Most nuclear forensics work is not continuous, so the majority of current practitioners must be occupied with other work for much of their time. Personnel in nuclear forensics participate in exercises, which are necessary to maintain competence, but exercises alone are not sufficient to occupy these personnel. It would be desirable if the other work in which personnel were engaged were complementary to nuclear forensics work. Fields such as environmental remediation, advanced fuel-cycle research, and nuclear medicine development and production are examples of relevant related work for radiochemists. Weapons designers mainly work on stockpile stewardship and on R&D to support that program. But for weapon designers there is still relevant nuclear forensics work outside of exercises, such as analyzing the predicted performance of unconventional improvised-nuclear-device designs, foreign weapons, underperforming known designs, and nuclear explosives detonated in urban environments.

To improve the personnel base for nuclear forensics, the committee suggests increased and sustained funding for nuclear forensics exercises, building and sustaining better connections with universities, and engaging workers from academia and industry with crossover skills on a reserve basis. The NTNFC has established the National Nuclear Forensics Expertise

Development Program to address some of these issues. This program needs to be sustained to have an impact on the nuclear forensics capabilities.

Recommendation 3: DHS and the cooperating agencies should implement a plan to build and maintain an appropriately sized and composed workforce.

To do this, the nuclear forensics program needs:

• synergistic professional work to engage program personnel when they are not working on nuclear forensics exercises and events;

• an R&D program focused on useful, challenging, and interesting goals;

• an exercise, evaluation, training, and learning program;

• collaborative relationships among the laboratories and an external community of experts and organizations to review and contribute to the program, to supply new personnel, and to assist the national laboratories in case of an event; and

• stable and sustainable funding.

Several of the facilities used for nuclear forensics are old and outdated, were not built to modern environmental, health, and safety (EH&S) standards, and are in need of replacement. The cost of constructing new facilities for handling special nuclear material² and highly radioactive material is large, so careful investment or replacement decisions are required, particularly if necessary redundant analytical capabilities are to be maintained. Funding for these investments is difficult to obtain today in part because a coordinated and integrated program plan has not yet been created.

Post-detonation nuclear forensics capabilities have been demonstrated in test and training exercises, but their operational readiness has not been tested in some realistic scenarios or no-notice events. With more realistic exercises (e.g., no notice, samples that are realistic outliers, real-time distribution of samples to the laboratories), managers will gain greater confidence in assessments and exercise participants will identify strengths, gaps, and deficiencies. This will enable DHS and its partner agencies to improve performance.

Pressure for increasingly stringent interpretation of EH&S goals has made it increasingly challenging to work with materials and equipment that emit radiation or are otherwise hazardous. This problem is particularly acute at DOE and NNSA facilities where concerns about health and safety can, in some cases, add a substantial time and cost burden to, or even preclude, the performance of tasks that are essential to national security. The full implications of these burdens for nuclear forensics are not known. Dealing with exigencies that might arise in such cases takes time that would be better spent on analysis, so it makes sense to work through predictable obstacles (EH&S restrictions among them) in advance.

Recommendation 4: DHS and the cooperating agencies should adapt nuclear forensics to the challenges of real emergency situations, including, for example, conducting more realistic exercises that are unannounced and that challenge regulations and procedures

followed in the normal work environment, and these agencies should implement corrective actions from lessons learned³

In post-detonation exercises, national laboratories have demonstrated that they can characterize nuclear debris and other forensic data and infer key design features of a variety of nuclear explosive devices. However, many of these characterization techniques are based on legacy procedures developed for analyzing U.S. and foreign nuclear weapons tests. Some of the techniques used in the past yielded results quickly but cannot be employed today because they would not meet contemporary EH&S requirements.

There are numerous opportunities for the United States to improve its technical nuclear forensics capabilities and performance. The top priorities are to increase analytic and operational capabilities in ways that reduce timelines and uncertainties in findings. The committee judges that marked reductions in timelines and uncertainties are achievable with reasonable efforts in R&D and operational improvements. In light of the cost-benefit tradeoffs inherent in the selection of a timeline, this selection has to be made by policymakers with input from the technical community.

The committee favors the development of an R&D effort that pursues both evolutionary improvements in the existing nuclear forensics methods and techniques as well as entirely new methods and techniques having far greater capabilities than those currently available.

In addition to building the expert personnel base, the program needs:

• Improved sampling procedures, especially for urban environments, informed by simulations of a variety of detonation scenarios;

• Improved laboratory techniques, including increased automation, improved techniques for sample preparation and laboratory analysis that extract more useful information from a spectrum of sample types and matrices, and techniques that more readily comply with EH&S requirements;

• Proper and defensible validation protocols for analytical methods;

• A better understanding of measurements that have been and could be made on nuclear materials production facilities and nuclear detonations;

• Improved techniques to assess uncertainties in sampling, laboratory analysis, and data evaluation;

• Validated and more complete databases of nuclear materials, facilities, and devices, including advanced querying techniques to extract information from incomplete and/or noisy data and find signatures and correlations among entries; and possibly

• New or newly deployed equipment for prompt diagnostics.

Modeling and simulation are used in modern scientific and engineering programs to guide experiments and examine the possible outcomes of events that cannot or will not be tested physically. The nuclear forensics program would benefit from increased use of modeling and

simulation for informing decisions about how to allocate resources, what experiments to perform, and where and how to sample in a post-detonation scenario. The needed work includes

• development and deployment of prompt diagnostic systems;

• simulations to better interpret prompt diagnostic measurements;

• identification of useful short-lived signatures and how to measure them;

• better planning for ground and air sampling, including simulations of debris fields and investigation of preexisting objects which, in the presence of an attack, can record signatures;

• development of tools and procedures for better sample selection;

• formalizing procedures for rapid sample distribution to and receipt by fixed laboratories;

• development of faster, more reliable sample preparation techniques;

• development of automated analytical techniques that meet modern EH&S requirements;

• simulations to explore signatures of nuclear-material-production technologies; and

• simulations of explosions of various postulated nuclear explosive devices, their effects, and resulting debris fields in the types of environments that are the most likely targets of a nuclear attack.

Recommendation 5: The laboratories involved in nuclear forensics should conduct research and development aimed at optimizing procedures and equipment to meet evolving program requirements. Modeling and simulation should play an increased role in several parts of the program, both for planning and for conducting analyses. This work should be funded at a level consistent with the overall implementation plan.

Nuclear forensics is a specialized subset of forensic science, a field that has developed over several decades and that continues to pursue properly validated standards and procedures. Forensic science today is moving to build an even stronger scientific base. There is an apparent tension between the requirements for forensic science that have evolved to support prosecution in a court of law, which is the usual context in which forensic science is employed, and what may be desired from forensic science to support decision making in a policy context. For example, the chain of custody of evidence is an established critical requirement in forensic science to support criminal prosecution, which may be difficult to sustain in a nuclear event. The chain of custody for a post-detonation sample may never meet the rigors required for a courtroom. However, this tension can be overstated. Several underlying principles have been recommended to guide forensic science.⁴ Those principles that are focused on ensuring a robust, reliable, scientifically sound analysis that provides well-characterized results, including uncertainties, apply equally to nuclear forensics and other branches of forensic science. While nuclear forensics procedures and standards will be developed for emergency or near-emergency situations, it makes sense for them to be rooted in the same underlying principles that are intended to guide modern forensic science. Many nuclear forensics techniques already have a strong scientific basis and so would fit well into a scientifically based forensic science framework.

Demands from the White House and Congress for immediate (even if preliminary) information in a post-detonation situation would be intense. In particular, there would be pressure to provide information that bears on attribution. At every stage of nuclear forensics, communicated results should be accompanied by an assessment of the confidence in the results and the implications for uncertainties arising from their use.

Recommendation 6: The nuclear forensics community should develop and adhere to standards and procedures that are rooted in the applicable underlying principles that have been recommended for modern forensic science, including calibration using reference standards; cross comparison with other methods; inter-laboratory comparisons; and identification, propagation, and characterization of uncertainties.

Even a casual observer of forensic science understands that databases are important tools for determining the possible origins and history of a material or an object: analysts compare the sample under scrutiny to a set of known samples. All nuclear material has history and it is a task of nuclear forensics to uncover that history to the extent possible.

Several U.S. agencies maintain databases populated with information that could be relevant to nuclear forensics and forensic analysis of nuclear incidents and events. Some agencies do not share their databases, or their databases are not readily accessible, which can be detrimental to performance of the program. The Nuclear Materials Information Program (NMIP) was created as the clearinghouse for information on nuclear materials inventories and characteristics. NMIP is now expected to link most but not all of the relevant databases.

Recommendation 7: DHS and the cooperating agencies should devise and implement a plan that will permit, under appropriate conditions, access to the relevant information in all databases—including classified and proprietary databases—for nuclear forensics missions. This means that, when queried, the responses to the queries will be timely, reliable, and validated, and will provide sufficient relevant data and metadata⁵ to enable analysts to use them.

International collaboration and data sharing are important. Many scenarios for nuclear forensics involve non-U.S. material, so data on non-U.S. material (including foreign military material) would be of great value. An international database of special nuclear and other radioactive material properties from sources all around the world would be a valuable resource and therefore is a worthy objective, but there are many challenges to achieving a reliable, comprehensive international database. The three greatest challenges are: (l) source bias (skewing of the data resulting from who is willing to participate and what data they are willing to share because of security concerns, political considerations, or proprietary restrictions); (2) unreliability due to intentional and/or inadvertent misinformation; and (3) the potential for inconsistencies between international and domestic databases that could make it more, rather than less, difficult for the United States to convince other countries of a U.S. conclusion.

The United States is encouraging countries to develop their own databases, but not necessarily to link them. This approach encourages other countries to focus on nuclear security and nuclear forensics, to include sensitive information in the databases they develop, to have the

ability to access that information in case of an incident, and to create entities with which the United States can communicate on these matters in case of an incident.

Recommendation 8: As the U.S. government organizes and enhances its databases and nuclear forensics methods, the Executive Office of the President and the Department of State, working with the community of nuclear forensics experts, should develop policies on classes of data and methods to be shared internationally and explore mechanisms to accomplish that sharing.

The United States should decide whether to share analytical methods to foster development of a broader international scientific base for conducting nuclear forensics. The United States should also conclude whether it would be useful to share its own analytical results to build international support for action following an event. These decisions should be made as soon as is practical, before the capabilities are needed to respond to an event.

For a decade, the U.S. government has considered a nuclear terrorist attack on the United States to be the most catastrophic threat the nation faces. Nuclear forensics is an important part of our response to that threat and important to our national security. It is not intended to provide all of the answers that decision makers desire, but to appreciate its value, one need only imagine the circumstances if the nation did not have a nuclear forensics capability in place when an interdiction or a detonation event occurred.

An impromptu nuclear forensics effort would be initiated. Such an effort would likely provide inferior and possibly misleading results on a longer timeline and with lower confidence levels. Key information might never be discovered. The intelligence and law enforcement communities would be asked to carry out their investigations without timely, reliable information about the nuclear materials or device design. Analysts would have no benchmarks by which to judge the quality of the information upon which decisions and actions are taken. This would make it more difficult to identify the perpetrators, especially their supporters or sponsors, and to inform the president of which countries may have been involved and which ones likely were not involved. Furthermore, even if intelligence and law enforcement were to successfully identify culprits associated with smuggling or detonating a weapon or material, they would have to provide supporting evidence for their conclusions, not just in a prosecutorial context but even in a national security context, domestically and internationally. Allies and adversaries alike must be persuaded, and that task would be made easier with physical forensic evidence provided by a robust program.

Important as nuclear forensics capabilities are, they are at risk and actions are needed now to sustain and improve them.