It’s great to be here. This is an event that Frank and Madeleine and the deputy secretary and others of us have been really looking forward to as a really important event. And by the way, this is the first of a double-header today in the sense that we are also hoping to see you in the afternoon on the Hill for our national security focused lab day. The first lab days have been really terrific events, lots of congressional participation. And obviously very appropriate today to kind of pair this up with the national security focused lab day.
I also want to just say that I think Frank, and Madeleine and others of the have really been providing terrific leadership. And we really have a very strong national security team right now going forward. I should also acknowledge and thank the Navy Heritage Center for hosting us today in this facility.
Now, I’m not going to go through names because, frankly, I see too many familiar faces here, which means of course we’re all 20 years older than we were when we started this effort back in the ’90s, and continue to see it through today. Besides the familiar faces, I think we can see some of the younger faces that General Klotz just referred to. I think we have quite a few of NNSA’s graduate fellows here today – Project on Nuclear Initiatives, Program for Emerging Leaders, some from the minority serving institutions that NNSA is serving. So I certainly would just reinforce Frank’s comments that we’re looking forward to you speaking at this event in the next 20 years. That would be great.
As Frank said, you all know we’re here to celebrate – kind of put an exclamation point on this 20th anniversary of President Clinton’s firing the starting gun on stockpile stewardship in the context of a commitment to the CTBT. We’ll come back to that. It was – obviously there was some bumps in the road and – that we will be addressing. But I think – coming back and looking back at this now, after my first go around the DOE track in the ’90s, that it’s a remarkable story that I think we will be drawing out today, combining science, national security, policy, politics, innovation, culture, skepticism, big institutions, fortitude, courage, and some degree of success. And it’s really remarkable, but of course we want to revisit this with a – with a forward-looking orientation.
And that is one reason why I really looked forward to being able to make these remarks. My fundamental story here is that what was done and started in the ’90s with science-based stockpile stewardship – I’m sorry, Frank; I still say SBSS, not SSB – it’s a story of incredible innovation. I’ll come back to this, but just the theme – the punchline – is nothing about this program was guaranteed to succeed. There was nothing off the shelf about it. Science and technology had to be invented to go into completely new domains. Obviously we’re here, in the context in which it’s largely worked. That innovation – again, I’m just finishing the punchline. I could sit down after this, but I’ll take more of your time. That innovation is what is critically important, not only about the nuclear security laboratories, but our laboratory system, to meet our nuclear security needs and our other critical mission needs. And that’s what comes not by trying to turn things on and off, but by making a sustained investment over many decades to sustain that innovation. That’s, in a nutshell, what I’m here to talk about. And I think that that commitment to innovation is something that I feel very good about seeing across our lab system. This is today a critical example of it, of a broader issue.
Now, Frank mentioned that President Clinton, on August 11th, 1995, issued his statement. A quote from there is: “I” – President Clinton – “am assured by the secretary of energy and the directors of our nuclear weapons labs that we can meet the challenge in maintaining our nuclear deterrent under a Comprehensive Test-Ban Treaty through a science-based stockpile stewardship program without nuclear testing.”
The directors got a little bit wobbly on that for a while, but we’ll come back to that. But they supported that statement.
I should also add that we would have liked to have done this on the 20th anniversary, August 11th, but August 11th was halfway through a certain 60-day congressional period for another example of science-based nuclear security. You all know what I’m talking about, and I’ll come back to that later on.
Anyway, the labs have met the challenge and drove science – and science and technology really to new heights. And I’ll say something about the program and the individuals.
Obviously there was the story before the story. But I’m not going to go back to Oppenheimer and Teller and the whole series of events over more of the half-century scale. But going to the ’90s – and, as Frank said, the collapse of the Soviet Union; the subsequent issues with Russia a few years later; end of the Cold War; some uncertainty in our missions, frankly. But the Soviet nuclear arsenal and institutions turned over to the new Russian Federation, with its enormous economic challenges in the early ’90s – and by the way, in the late ’90s, too, which influenced very strongly what we could do.
President Bush senior halted production of any new nuclear warheads. And this was followed by President Clinton’s direction in 1993 to the Department of Energy to explore other means to maintain the confidence – and that’s a quote – “explore other means to maintain the confidence in the safety, reliability and performance of our own weapons.”
So the question is of course, what were the other means, and what do you mean by confidence? It was clear that this was a sea change for the enterprise, which by that time had conducted over a thousand nuclear tests; had fielded 70 types of weapons on platforms from artillery shells to ICBMs, small to very, very large. The sudden end – kind of the music stopped in the musical chairs game – the sudden stop of the Cold War really meant a complete rethinking of 40-plus years of practice essentially overnight, and a starting place at the gate that had a stockpile that certainly was not designed or optimized for these changed conditions.
Responding to that, our curmudgeonly friend down here in the front row, Vic Reese had just been confirmed as assistant secretary of energy for defense programs. Welcome to your new job, by the way – it’s totally changed. Vic met with the lab weapons program managers at that time. Familiar names – John Immele, George Miller, Roger Hagengruber – and their kind of chief weapons scientists – John Browne, Dick Fortner, Gerry Yonas – and basically, over a couple of days of brainstorming, laid out this new strategy around SBSS. And by the way, Charlie was living through those years, as well, at DOE.
The concept was very elegantly simple. It was only a question of doing it. The nuclear explosive process was to be broken down into all of its major component processes. Those processes were to be experimentally investigated to measure in sufficient detail, to validate computer simulations that would kind of pull these data together, and test designer understanding with each of these validated simulations assembled to simulate the entire nuclear weapons explosive process, and then the result compared with previous underground testing data.
But again, a few problems. None of the experimental tools existed. And clearly, the tools that had to be developed – first of all, the experimental tools – were going to have to explore physical parameter spaces, temperature/pressure that were certainly unknown in the laboratory; and then, on the computational side, looking first at a required hundred-teraflop computational capability – a factor of 10,000 faster than the then-fastest computer, and really a factor of a million greater than what was typically used by a designer in those – in those days. And as I’ll come back to, it’s not like, OK, let’s just get 10,000 of those and put them together. A literal paradigm shift had to be put into place.
There was certainly little experience on things like how nuclear materials – plutonium – change properties with age; how those changes would affect safety, reliability and performance of the – of the weapons system. And while the stockpile was kind of viewed as in being in decent shape, certainly in the context of the traditional paradigm, but we were also at a stage where this un-optimized stockpile really had a lot of weapons reaching their maturity, shall we say.
So let’s face it, there was a great deal of technical uncertainty embedded in a policy debate ranging from those who were looking for a rapid reduction in the U.S. nuclear weapon enterprise and with it a budgetary peace dividend, to those who were skeptical of any Russian or Chinese political reform and pushing for continued testing and a modernization of the stockpile and its aging industrial base. But the labs pushed on to the task of designing the new system, designing the new facilities, building those facilities in conjunction with the program, which had non-nuclear explosive testing as part of its activity. Whether it’s NIF at Livermore, DAHRT at Los Alamos, MESA at Sandia and other facilities, including subcritical experiments at the Nevada Test Site, these all kind of moved from nice to have and we can kind of take our time getting there to must have, and must have it in a very, very timely fashion.
I do see, I think, Gil Weigand, there as well. Curmudgeon. I won’t characterize the other thing – (laughter) – I say about Gil, other than to say we had a hell of a lot of fun moving this forward. But in particular on the ASCI – on the Accelerated Strategic Computing Initiative – and by the way, now the president’s order of a few months ago has – just replaces “Accelerated.” Now we have the National Strategic Computing Initiative, which will take us to exoscale. But Gil obviously played a huge role in the ASCI, a truly integrated lab and manufacturer program to meet the goal of that factor of 10,000 within a decade, but of course also providing important and useful results all along the pathway to that hundred-teraflops goal.
Again, I’m going to come back to this in the context of innovation. But once again, little of this could really be planned all the way to the end, I think, until we got into the program and started to recognize the challenges, and then overcome them one by one. This certainly could not be done by an active genius, by one person, but really by the cooperative effort of huge teams of scientists and engineers, principally at our labs, working together in very considerable multidisciplinary teams.
Now, with this general plan outlined the president made that statement that I referred to earlier, with the assurance of the lab directors given the state of our scientific capabilities at that time. In his 1996 testimony to Congress, then-Sandia Director Paul Robinson stated the challenge pretty succinctly. And again, I quote: “The commercially available and laboratory technologies are inadequate for the stockpile stewardship tasks that we will face in the future. Another hundred- to thousand-fold increase in capability from hardware and software combined will be required. Some aspects of nuclear explosive design are still not understood at the level of physical principles.” I think that pretty well summarizes this state of affairs as we launched into realizing SBSS.
So Robinson and his fellow lab directors established that science-based meant two things: pursuing the science and tools needed to fill the gaps in our knowledge, and being very clear about the uncertainties.
Now again, I came to the department for the second Clinton term, soon after the United States was the first to sign the CTBT, and I can remember sitting not very happily with Secretary Richardson and the lab directors at the Senate witness table for the CTBT hearings in 1999. The treaty faced serious skeptics, and I’m going to say, quite honestly, not unreasonably. For example, a letter from Henry Kissinger, Brent Scowcroft and John Deutsch – which you might call members of the foreign policy realists camp – captured the mood of the debate. Again, a quote: “But the fact is that the scientific case simply has not been made that over the long term the United States can ensure the nuclear stockpile without nuclear testing. The Stockpile Stewardship Program is not sufficiently mature to evaluate the extent to which it can be a suitable alternative to testing.”
That brought along another 30 votes in favor of the CTBT – that’s a joke obviously. So clearly it was not ratified at that time, and the hearings were something – I think that the technical description is they were a debacle. And the lab directors, while I would say expecting success in the science-based stockpile stewardship, were unwilling to guarantee it until the tools were there and the program demonstrated. And frankly, to me that is a reasonable position.
But the program went on, providing the technical basis for the program. By going through this really extraordinary scientific process, it was determined that the lifetime of the current warheads could be extended for at least another generation by essentially a rigorous refurbishment, and that no nuclear warheads were going to be required to meet the Navy or Air Force specifications.
One of the major technical issues was estimating, for example, the lifetime associated with plutonium aging, and whether a large new plutonium pit factory would be required promptly. We all know how that’s been resolved. But last year I asked NNSA in the labs to make a detailed, bottom-up review asking the question – suppose the Senate decided to hold CTBT ratification hearings again; what could you say about our confidence in the stockpile? Suppose you were now at this new witness table in the no-testing regime. And clearly, from my perspective, the results were quite gratifying after our lab directors led this activity. Every science-based stockpile tool that had been planned had been built, was operating, delivering results, and in many cases well beyond the original expectations. Certainly the ASCI program that I referred to earlier met the “impossible” goal of a hundred teraflops by 2004. It’s “impossible” in quotes, because in the decade since, we’ve gone another factor of a hundred or something, up into the tens of petaflops level – 25 petaflops, let’s say.
Working in partnership with the DOE Office of Science labs – and I want to emphasize that is something that we also insisted upon in the late ’90s, that there would be a huge payoff from using what defense programs was doing on the science and energy side, but there would be a very important feedback loop in terms of bringing new people and new capabilities to things like the kind of new algorithmic developments that one needed for chemically reacting flow problems, which appear in many contexts like supernovae, for example, as one of the examples that was followed. So I think that was another very important thing. And as many of you know, today the Office of Science is, in fact, helping to push those frontiers. Especially, as Frank said, we push towards the exoscale in the context of the National Strategic Computing Initiative.
So today we can do those calculations at an unbelievable level. But also physically we can – with these capabilities – trace nanoscale instabilities that are important for the physical understanding of the weapons process. And these are tremendous, tremendous anchors that allow us in the nuclear security mission as well to go beyond the stockpile requirements to things like nuclear forensics, threat detection, hostile environments for weapons performance, but also to things like the formation of the universe.
NIF I mentioned earlier, another example of flagship facilities reaching higher densities, pressures, temperatures than any other facility in the world. For example, some applications exceeding 100 million degrees – the kinds of extreme environments that we need to for looking at the stockpile, but at the same time opening up whole new frontiers of high-energy-density science. Pressures, hundreds of megabars, gigabars. Again, NIF addressing areas such as high-pressure strength equations of state, hydrodynamics, dense plasma effects – many of the key scientific areas underpinning our continued confidence in the science-based stewardship of our nuclear weapons stockpile. I’m sorry if I’m going on and on on this, but I just think it’s really important to understand what this meant in terms of science.
So, with many of the successful innovations that stewardship has produced, the directors of our national security laboratories have been able – first of all, to certify the diminishing stockpile annually since that time in 1995. And the lab directors today now state that they certainly understand much more about how nuclear weapons work than during the period of nuclear testing and, if we had continued the paradigm, what we would understand today in terms of the physics of these devices. So the underlying theme of deterrence is confidence, and certainly critical that this confidence flow up to the president and to the Congress in terms of stockpile performance.
Now, again, I just want to go back one last time to this question of the innovation. And as I said earlier, there’s nothing off the shelf about having achieved these goals. And you know, I think today’s a time to remember that we should not take this for granted.
Again, if I take ASCI as an example, it wasn’t just about buying 10,000 of something that existed. It was about pushing the paradigm of parallel architectures. It was about innovation in the business model, if you like – the way the department, the labs, and the major computer vendors came together to advance this. And I think that as we now look to this next push to exoscale, we are looking again at a kind of a paradigm shift in how this is going to be achieved. This is not going to be just more of the same, from the big data requirements, to the integer operations, to things like energy management. A hundred times a few megawatts doesn’t work for a computer, except for those who are such big SMR advocates that they would like one per computer.
So, anyway, this is going to be the same kind of innovation challenge that we need. We’re now hearing about MaRIE and can we reach a whole new generation of imaging at the very fine materials from very small ranges to mesoscale in terms of materials dynamics under extreme conditions? Yesterday we had another wonderful event, which was the awarding of the Fermi prizes to Claudio Pellegrini and Chuck Shank. And again, same kind of story – incredible new capabilities, from X-ray FELs under construction at SLAC, to Chuck Shank is we could say the inventor of femtosecond in terms of – terms of lasers. And again, these are the kinds of capabilities that are just really extraordinary and important.
I mentioned earlier our disappointment in ’99, in terms of CTBT discussion. Today I would argue we certainly have a much stronger case to make with not only SBSS but also, we should remember, the advancements in international monitoring and verification over the past 17 years, to which the United States and the labs have made a number of significant contributions. I visited the Preparatory Commission for the Comprehensive Nuclear Test-Ban Treaty Organization a couple years ago and was very impressed with the team of international experts in nuclear explosion monitoring and verification, supported by experts from 183 state signatories. The verification regime, which was just a concept two decades ago, is now close to being a complete international monitoring system supported by the International Data Center.
The system has effectively demonstrated its capabilities, detecting and helping states identify the three declared nuclear explosive tests in North Korea over the past several years. But also other applications – the 2004 Indian Ocean earthquake and tsunami; 2011 Fukushima nuclear crisis – also showed how the international monitoring system can serve important non-verification-related purposes as well.
Nearly 90 percent of the planned international monitoring system stations are already certified or installed, with plans for additional stations. And I would note that a great deal of technology used by the stations and in the radionuclide laboratory originated from DOE lab experts in seismology, infrasound analysis, hydroacoustics and radiation detection.
I want to close by talking about the role of DOE’s science-based nuclear enterprise more broadly, and particularly to bring up the context of the Iran nuclear deal – as I said earlier, another very important aspect of science-based security. The results of the investment in these areas – in science-based nonproliferation programs – the results of those are really nothing short of heroic: securing nuclear materials globally, including removal of all HEU from 28 countries; development of novel technologies to discover foreign nuclear weapons development activities, detect nuclear detonation, strengthen monitoring and verification, characterization, detection and defeat of the range of nuclear or radiological devices potentially available to a rogue state or terrorists.
And many of you know that one of the features of the Iran agreement and the Iran verification thing is that technologies that have been built up over 20 years are going to be an important part of the new safeguards. In fact to tie back to my last go-round, one of the things that was developed at a couple of our labs was real-time monitoring of enriched uranium streams. That was in the context of the Megatons to Megawatts Program. Well, now that’s going to be critical in monitoring the enrichment activities in Iran. So these things really are the result of sustained investments.
Now, in late February, the president asked me to join Secretary of State Kerry in the P5+1 Iran negotiations, taking the role of lead U.S. negotiator for the nuclear dimensions of an agreement that would verifiably prevent Iran from having nuclear weapons or detect violation of the agreement in a timely way if Iran were to pursue a weapons pathway. There was the inevitable question asked in the popular discussions about, roughly speaking, what the hell’s a Secretary of Energy doing in a sensitive high-stakes diplomacy on an issue of critical importance to the United States and our allies, especially in the Middle East? But this audience knows the answer transparently: it’s because the Department of Energy was always supporting the negotiation. Who else is going to provide the nuclear expertise that underpins a highly technical agreement? And actually seven of our laboratories and two of our nuclear security sites were intimately involved, bringing complementary skills and capabilities to the task.
And that to me reinforces the earlier statement I made: we need to keep sustained, steady investments in these capabilities if we are going to be able to respond to the kinds of challenges that come up, probably I would say too often. But some of you may have seen that one of the now very specific responsibilities that we will have, among many, in the implementation phase, we published on Sunday, which was adoption day for the agreement, a statement of intent worked out between the United States, China and Iran, where the United States and China will co-chair the working group for the redesign of the Arak reactor so that it is, roughly speaking, not a plutonium factory. Well, obviously, again, that expertise is going to come from our laboratories.
So, in concluding, this 20-year anniversary of President Clinton’s statement really is an appropriate time to look back at where we’ve come – looking at the innovation that was certainly an aspiration in 1995 and today, the fruits of it are very clear. The Moore’s Law speed limit apparently could be violated in terms of computing. But also the experimental facilities going into just extraordinary parameter spaces, as we discussed earlier. And also with the spin-outs and the interplay back and forth between other capabilities like genomics.
And another example, by the way, is how the whole genomics initiative got kicked off. And now, today, it’s no secret. We had a workshop on Friday at Argonne on the brain. Why? Because NIH has come to us and said, we need the capabilities of these laboratories for a problem of – it’s not, quote, “in our mission space” directly, but we have the capabilities that are needed for this – for this program of tremendous national importance.
So that’s really the message, is the innovation system that we must sustain on today’s theme. It certainly gives me a lot of confidence that, you know, maybe it’s time to revisit that discussion of 17 years ago, 20 years ago, the CTBT debate, in a very, very different context in terms of where we are scientifically. And that’s certainly a discussion that I, for one, would kind of welcome to be engaged in, and to talk about our readiness for such a move in a way that we were not, frankly, in the late ’90s.
So thanks to all of you – and there are many, many here – who have been key contributors to this progress, and just keep it up for the next few decades.
Thank you.