Testimony Before the Senate Committee on Governmental Affairs, International Security, Proliferation, and Federal Services Subcommittee on October 27, 1997

Prepared Statement of: Victor H. Reis, Assistant Secretary of Energy for Defense Programs

Thank you, Mr. Chairman for the opportunity to testify before you today on the Stockpile Stewardship Program. This program is fundamental to our national security under a Comprehensive Test Ban Treaty. I'd like to begin with a brief history of stockpile stewardship, tell you what it is, give you its current status, and then answer your questions. In addition to my written testimony, I would like to provide the subcommittee with a recently published overview on the program, and if you wish, submit it for the record.

The Stockpile Stewardship program began in July 1993 when President Clinton announced he would continue the moratorium on nuclear weapons testing and seek a comprehensive test ban treaty for nuclear weapons, a goal that has been sought since President Eisenhower. In August of 1995 President Clinton announced his intention to seek a "zero yield" CTBT. He included as part of his announcement, six safeguards that would accompany the treaty. The first of these was that we will conduct a "science based stockpile stewardship program." The Senate Start II ratification text in January 1996 also commits the U.S. to a "robust Stockpile Stewardship Program."

President Clinton signed the CTBT in September of 1996, and on September 22 of this year he submitted it to the Senate for approval. As part of the submission, the Administration committed to fund stockpile stewardship at about $4.5 billion in FY1999 and to use FY 99 as a baseline for future funding. This does not include funding for construction of a new tritium production source. Thus, stockpile stewardship -- which is essential to maintain our nuclear deterrent -- also underpins the nation's nuclear arms control policy.

As President Clinton stated in August of 1995:

    "I am assured by the Secretary of Energy and the Directors of our nuclear weapons labs that we can meet the challenge of maintaining our nuclear deterrent under a Comprehensive Test Ban Treaty through a science based stockpile stewardship program without nuclear testing."

Thus, Mr. Chairman, within the U.S. national security framework, the specific task of stockpile stewardship is to maintain high confidence in the safety, reliability, and performance of the nuclear stockpile, indefinitely, without nuclear testing. And part of this task is to maintain the capability to return to testing and production of new weapons, if so directed by the President and the Congress.

So, what is the program, what are the risks involved, and how do we plan to mitigate those risks?

The stockpile stewardship concept is simple. Each year representative samples of each type of weapon are returned from the active forces to the plants and labs, disassembled, examined, tested and analyzed for defects, much as you would go for an annual physical or take your car into your local automobile mechanic. If any defects are found, their effect on performance safety and reliability is assessed, and if that effect is deemed significant, the defective part is remanufactured and replaced. Like the battery or spark plugs in your car, some parts we know will require replacement, and these are replaced at regular intervals. That's it. It sounds simple enough.

Unfortunately, while a modern nuclear weapon has about as many parts as a modern automobile, it is much more complicated. Many of the parts of a nuclear weapon are made from very special materials - plutonium, enriched uranium, tritium - which radioactively decay, and change both their properties and the properties of other materials within the weapon.

Nuclear weapons are designed and manufactured to extraordinarily rigid standards, both to enable huge amounts of explosive energy to be packaged in relatively small containers, and to maintain phenomenal safety standards. A nuclear weapon, less than the size of a small desk, will have the explosive power to completely destroy a modern city, and yet it must be able to survive the worst kind of accident you can think of with less than a one in a million chance of exploding. This level of performance and safety must be maintained throughout the weapon's lifetime, even as it ages and changes.

While we can expect that aging will cause the defect rate to rise - just like it does in both humans and cars - we can't go out and buy a new warhead model - there is no new warhead production, and some of the old factories are out of business. Moreover, the weapons designers who have had experience with nuclear explosive testing are also aging, in about ten years most of them will have retired. This means that about the same time all of the weapons reach the end of their design life, we will no longer have anyone on the job with direct test experience!

Despite these challenges, people from the weapons laboratories, the production plants, and the federal establishment involved in stockpile stewardship have testified, and will so testify, that we can do the stockpile stewardship job. We believe we can maintain the safety and reliability of the nuclear weapons in the stockpile indefinitely without underground testing and keep the risks to manageable levels.

How do we expect to do this?

First of all, we start from a solid position. The current stockpile has been well tested, is in very good shape and is well understood. We have an extensive data base on each of these weapons, and we have a cadre of experienced designers, engineers, scientists and technicians that can, with confidence, certify the safety and reliability of the current stockpile.

Now, since we cannot do a complete test of a nuclear explosion, we conceptually divide the explosion into each of its parts and test and analyze each of these separately, much as you would test the ignition system, the cooling system, and the brakes on your car. We then put the whole thing together into a computer calculation - a simulation - to see if the resulting performance is within its specification. Each part of the simulation must predict the results of each of the separate tests, and where they exist, be consistent with data from previous underground nuclear tests. Let me give you some very simplified examples of how this works.

Some of processes are relatively straight forward to simulate. The first part of the nuclear explosion sequence is to send the right electrical signal to the right place at the right time. We can test this exactly by flight testing actual weapons with inert mockups of the nuclear components.

We can do a good job of testing the first part of the nuclear explosion, the implosion of the plutonium pit, but we do not use actual plutonium - it would go off if we did - and we can measure a number of important features by taking x-ray pictures during critical parts of the experiment. We can then compare these pictures with calculations and with previous actual underground nuclear test results. But current radiographic systems will not be sufficient to measure the effects of potential defects in an aged pit, so we are building a new x-ray machine - the DARHT - which will look at the shape and size of an imploding pit model from two different directions and with much better resolution.

Beyond obtaining x-ray pictures of imploding pit models, however, we will no longer experimentally simulate a nuclear explosion, but instead use experimental facilities to obtain conditions that occur during such an explosion and then use the results of these experiments to check computer calculations. For example, we are investigating the way old plutonium behaves when subjected to the high pressures of an implosion, through subcritical tests at the Nevada Test Site, and we expect to be able to generate the conditions of temperature and pressure of nuclear explosions with lasers at the National Ignition Facility. These, and other experimental facilities that are on line, under construction, or in the planning stage, will give us a set of tools sufficient to investigate and help understand anticipated problems in the stockpile.

As I mentioned previously the experimental information is tied into the assessment process through computation, or more precisely, numerical simulation. But we know that the level of computation needed to effectively simulate effects of aging or a remanufactured part is much, much greater than that currently available, so we have begun a computation development program - the Accelerated Strategic Computing Initiative - in parallel with the experimental program. There is no point in doing elegant experiments if you can't interpret the results in terms of nuclear weapons safety and reliability, and there is no point in doing simulations if the computer codes cannot be grounded in reality. You need both, as well as returning to the archives to match the new techniques with the data from underground nuclear tests.

It is this troika of computer simulation, experiments, and previous nuclear test data that provides the complete tool box for the assessment process. Building this assessment "tool box" in time to train the new cadre of scientists and engineers is critical to the stockpile stewardship program.

This leaves remanufacture - we know now we will have to remanufacture and replace some parts, and are already doing so. We know that eventually we will have to replace just about every part in every weapon -- that's the idea of stockpile life extension. But to create these new parts we cannot rely on the cold war production complex that produced some tens of thousands of nuclear weapons. We are establishing a production complex that is much smaller, much more flexible, and much more environmentally sensitive than the production complex it replaces.

We must use every applicable modern manufacturing technique; the best that U.S. industry can offer. We must understand the details of the manufacturing processes with sufficient precision, so as not to introduce new defects into a remanufactured system. The key here is model - based manufacturing - similar to that which created the Boeing 777 and is being applied today by much of U.S. industry. Thus, around half of the stewardship program is devoted to producing current replacement parts, and to planning and modernizing our production complex to match the new job. We envision a complex of approximately 1/5th the size of the cold war complex, but one that can return to higher levels of production if the need ever arises.

While we do not expect to need additional supplies of enriched uranium and plutonium, there is one nuclear material which we know we will have to produce: tritium - a radioactive isotope of hydrogen that is required for every modern nuclear weapon.

Tritium decays fairly rapidly; approximately 5% is transformed to helium every year. The last tritium that was produced in the U.S was in 1988, but with the end of the cold war and the reduction of numbers of nuclear weapons, we have had large amounts of excess tritium. This excess has been used to make up for the decayed tritium in the current stockpile, but eventually this will run out. Based upon current estimates we must produce tritium by 2005 to support a START I nuclear stockpile. After a number of years of analysis and changing requirements we are down to two approaches for making tritium - using an existing commercial light water reactor or using a newly developed accelerator. The DOE will select a primary source for tritium production as soon as possible in FY 1998.

So in a nut shell, that's stockpile stewardship - maintaining the stockpile without testing - surveillance, assessment, remanufacture - tritium, labs, and plants, - a program that must develop a new generation of technical experts before the current generation retires.

Why do we think we can meet this challenge, and what are we doing to manage the risks?

First, let me reiterate that we start from a solid base. The current stockpile is well tested and well understood. The designers and engineers who built them are available and are active. Indeed they are the ones who are creating the stockpile stewardship program. They are the ones who are working on the stockpile now, and are helping to train their successors.

Second, we have laid out a plan for the stockpile stewardship program --- weapon by weapon, part by part, that projects the tasks that are required to maintain the stockpile over the next ten years, and beyond. We have concurrence on this program from the Department of Defense, and the Joint Chiefs, and the administration has committed to fund this program and all its parts.

Third, as one of the conditions for ratification, Safeguard F, the President requires us to annually certify, to him directly, the safety, reliability and performance of each weapon type. This is done by the Secretary of Defense and the Secretary of Energy on the advice of the Nuclear Weapons Council, the Directors of the nuclear weapons laboratories and the Commander - in - Chief of the U.S. Strategic Command. (If a high level of confidence in the safety or reliability of a nuclear weapon type which the two Secretaries consider critical to our nuclear deterrent could no longer be certified, the President, in consultation with Congress, would be prepared to withdraw from the CTBT under the standard "supreme national interest" clause in order to conduct whatever testing might be required.)

Fourth, we have a back up. Safeguard C, requires us to maintain the Nevada Test Site in a state of readiness, and the subcritical and other experiments conducted there help keep the people sharp and ready.

Fifth, Safeguard B states that ratification is conditioned on maintaining the vitality of the nuclear weapons laboratories - Los Alamos, Lawrence Livermore and Sandia National Laboratories. Mr. Chairman, those labs are among the best in the world - in my opinion they are the best in the world - and they are better now than they were four years ago because of the enthusiasm and vigor with which they are attacking the stockpile stewardship effort. History tells us that great labs need great missions, and stewardship is just such a mission. Our DOE labs will get even better because they will attract the kind of people who are drawn to solve tough problems of national importance.

Sixth, we are doing stewardship now, and doing it successfully. It has been five years since the last underground nuclear test. We are just completing our second annual certification. We have modified the B61 bomb and seen it enter the stockpile to replace the aged B53 bomb. We have initiated a number of new experimental tools, and our computation program has developed the world's fastest supercomputer - by a factor of three. And we have solved some problems by using stewardship tools that in the past would have likely required nuclear testing. We have literally done hundreds of experiments that increase our understanding of nuclear weapons. We have safely dismantled over nine thousand nuclear weapons since the end of the Cold War, have produced numerous parts, on time, while continuing to downsize the complex. This is a system that works, and not just at the labs but also at the plants: Oak Ridge Y-12, Pantex, Kansas City, Savannah River, and the Nevada Test Site.

So let me finish by getting to the essential question: Do I have confidence that stockpile stewardship will work, can we maintain the nuclear weapon stockpile, without testing, ten, twenty, thirty years from now?

My answer now is an (almost) unqualified yes.

The source of my optimism lies not in the immortality of the current stockpile of weapons - though in truth they are truly technological marvels - but in my faith in the integrity, courage and competence of the people in our weapons labs and production complex. They are the men and women that designed and produced the weapons that ended World War II and kept the Cold War cold. They have put together a program that is comprehensive, coherent and robust. They believe, and I believe, they can do the job, by first and foremost maintaining and supporting the institutions to do the job. I have confidence in them - their integrity, their competence and their overriding dedication to their mission. If we give them the tools that they need, and stick with it, we can manage the risk. In the end this is not an issue of technology but an issue of courage and will and persistence, and if we have the courage and will and persistence, we will not fail.

Thank you, Mr. Chairman, and I would be glad to answer any of your questions.