The Clinton administration has embarked on the most sweeping overhaul of the United States nuclear weapons program since the dropping of the first atomic bomb more than a half century ago.
With the US committed to a global ban on nuclear tests, the plan calls for using computer simulation and other advanced techniques instead of underground blasts to ensure that the atomic arsenal remains potent and problem-free well beyond its intended lifespan.
The program also seeks to preserve, at much reduced capacity, the nation's ability to design and build new warheads by maintaining a limited scientific and production base. The US stopped making warheads in 1989 after closing parts of its once sprawling nuclear-weapons "complex."
But even as it was quietly approved on Dec. 19 by the US Department of Energy, the Stockpile Stewardship and Management Program (SSMP) was mired in controversy. Despite a price tag of $40 billion over 10 years, even some of its managers agree there is no guarantee that all the technical barriers can be overcome. Nuclear critics, meanwhile, argue there are less expensive ways to maintain the US nuclear arsenal.
The program may also affect how US allies - as well as Russia and China- deal with their nuclear stockpiles in the future.
The uncertainty over the technical feasibility of the program has raised grave concerns, particularly among conservatives, over whether the US will be able to maintain a credible nuclear deterrent - a pillar of national security policy.
Though the danger of a nuclear holocaust receded with the end of the cold war, most experts agree that absent global disarmament, the US must retain a reliable atomic arsenal to confront potential threats, such as renewed rivalries with Russia or China.
But the long-term effects of age - corrosion and decay - on the 6,000-odd components that make up US nuclear bombs are unknown. The US used to replace warheads long before the end of their design lifespans and routinely detonated weapons taken from the stockpile to detect problems and confirm fixes. Tests were also used to verify new designs.
The US unilaterally halted tests in 1992. On Sept. 15, it joined other United Nations members in approving the Comprehensive Test Ban Treaty (CTBT), which the administration regards as a major tool in preventing countries like Iran from developing nuclear weapons.
The SSMP will attempt to replicate in the virtual world what the CTBT prohibits in the real world. But the challenges are formidable and success is uncertain.
The DOE must build at the three nuclear weapons laboratories five experimental facilities, including the world's most powerful laser device, that scientists hope will let them peer deeper than ever into the hearts of minute atomic reactions. Data from the experiments would be crunched to simulate full-scale nuclear blasts. To do that, scientists must develop a computer 10,000 times more powerful than the fastest existing machine, a 1 trillion-operations-per-second computer unveiled by DOE and Intel Corp. last month.
"At one level, this is like maintaining your car," explains George Miller, associate director of national security programs at Lawrence Livermore in California, one of the weapons laboratories. "You have confidence it is going to run by turning it on. But we're not allowed to do that any more. We have to run it artificially."
"That capacity does not exist right now," says Dr. Miller. "We are embarking on a program that has not been proven to work." But he adds that the US has sufficient time and expertise for the requisite research and development.
The US nuclear arsenal - which at its cold war high numbered more than 20,000 warheads of 25 different types - is being cut by agreement with Moscow. It now numbers about 9,000. The average age is about 13 years. By 2006, the stockpile will comprise 3,500 weapons if Russia has ratified the START II agreement. They will have an average age of more than 20 years, the planned lifespan of most warheads.
But by that time, experts believe they will have mastered the breakthroughs required by the SSMP. "We have some time because the stockpile is in pretty good shape right now," says Miller.
Physics aside, there is also a question of whether there will be sufficient funds to carry out the program. It was only after President Clinton assured them of full funding in 1995 that the weapons laboratories signed up to the SSMP. Since then, however, DOE has projected a $4.5 billion shortfall between fiscal 1997 and 2002.
"We have no chance of succeeding without adequate financial support," Dr. C. Paul Robinson, head of the Sandia National Laboratory, in N.M., warned a congressional hearing last March.
The $40 billion budget is the "minimum" necessary for the program, researchers assert. More will be required should scientists determine that for the SSMP to succeed, four additional high-energy experiments will have to be built.
Administration officials and program managers readily concede they are sailing into uncharted seas. Still, they are optimistic that given adequate resources, they will meet the scientific challenges. Besides, they add, the CTBT leaves the US no alternative.
"We all have our reservations and our concerns," says David Crandall, the DOE official overseeing the National Ignition Facility (NIF), a $1.8 billion cluster of 192 lasers to be built at Lawrence Livermore. "But we are confident we can do this job."
Others are not so sure. Serious misgivings prompted the Joint Chiefs of Staff to insist that the military be allowed to conduct an annual review to certify that the SSMP is on track and the stockpile secure.
Skeptics, including weapons scientists, insist there is no substitute to test blasts for ensuring the integrity and reliability of warheads. Without explosions, they argue, confidence in the stockpile will suffer and US security will be jeopardized.
"It is impossible to maintain a stockpile with the same standards of confidence and reliability and performance that we had in the past," says Jim McNally, a former weapons designer who was also involved in shaping US nuclear-weapons policies.
His concerns go beyond the CTBT's prohibition on testing. He and other critics believe the US may already have lost the capability to design and build weapons should new international instability prompt the White House to reactivate production in the "supreme national interest," as the president has put it.
They argue that to resume production the US must have the necessary facilities and sufficient number of experts steeped in the arcane knowledge that can only be gained from test explosions.
But, they continue, three key facilities were closed in the late 1980s before other parts of the complex were ready to assume their functions. The surviving plants are also in need of modernization.
Furthermore, they say, large numbers of veteran experts have retired or been laid off, preventing them from passing knowledge to others. Nor have they been replaced in sufficient numbers, a problem the DOE hopes to stem by attracting fresh talent with the new SSMP experiments.
Concerns over the shortage of experienced experts were expressed by officials at Lawrence Livermore in a 1995 study.
The study, speaking on behalf of all three of the nation's weapons laboratories, questioned "their ability to maintain US capability in nuclear weapons research and to maintain the skilled work force necessary for this work."
"Without the prospect of future stable funding, both recruiting and retention of highly skilled personnel will become increasingly difficult," the study said.
Such concerns are being examined on Capitol Hill in advance of the submission by the administration of the CTBT for Senate ratification within the next few months. Some majority Republicans may oppose the treaty, citing concerns over its impact on the safety and reliability of the nuclear arsenal.
In a preview of anti-CTBT arguments, Rep. Floyd Spence (R) of South Carolina, chairman of the House National Security Committee, released a report just before the Nov. 7 presidential election that denounced the SSMP as "erosion by design."
"Maintaining the safety and reliability of US nuclear weapons in the absence of nuclear testing poses serious technical and technological challenges and may even prove to be a 'mission impossible.' " says Rep. Spence.
Antinuclear groups, for their part, believe there are far cheaper ways to maintain the stockpile than building platinum-priced experimental facilities at the weapons laboratories.
Nuclear pork barrel?
They also allege that the facilities, particularly the NIF, were added to the program by the administration to induce the laboratories into supporting the SSMP. The DOE denies the allegation.
The Washington, D.C.-based Natural Resources Defense Council plans to file suit in coming days to block implementation of the SSMP. It claims that the DOE failed to comply with a federal environmental law that requires it to examine "reasonable alternatives" to the program.
Among other things, it says that in approving the SSMP, the DOE approved unnecessary, duplicative experiments, underestimated environmental and public health costs, and failed to consider "reasonable policy alternatives," such as a progressive elimination of nuclear weapons.