The Sky, Indeed, May Be Falling

While societal disruptions are bound to occur in every country at the turn of the century, the most serious and unforgiving aspect of the millennium bug could be a catastrophe at any of the world’s 433 nuclear power plants or an accident involving control of the 4,400 nuclear warheads that are maintained on hair-trigger alert in both Russia and the United States.

Yet, at a White House meeting I attended recently with John Koskenin, the head of the president’s Y2K committee, representatives of the Nuclear Regulatory Commission and the Department of Defense and four independent experts, the government representatives dismissed concerns that were raised while providing no substantive basis for confidence that we do not face potentially irregular, and possibly serious, nuclear accidents on or after Jan. 1, 2000.

Nuclear power plants are dependent upon an intact external electricity supply to maintain the circulation of about 1 million gallons of water per minute to cool the radioactive core and also to keep the spent fuel pools cool. If a section of the grid goes down, the approximately 100-ton fissioning uranium core in the affected reactor will melt within two hours if the two back up diesel generators--whose reliability has been estimated at 85%--fail.

Unlike the reactor cores, most of the spent fuel pools, which hold four to five times more radioactivity than the core, have no back up power supply nor containment vessel, and thus could melt within 48 hours if the reactor has been recently refueled; if not, they would melt within two weeks without cooling water. Twenty-six U.S. reactors are scheduled for refueling before Jan. 1.

While Koskenin admitted the possibility of random power outages in the U.S. electricity grid, he did not address the issue of the precarious back-up generators nor the fact that the NRC requires only one week of diesel fuel at each reactor site, even though local power outages could last longer.

The NRC representatives told us that contingency plans are in place, which we assume includes evacuation in case of a nuclear accident, and said that they are arguing with the Federal Emergency Management Agency about which agency will pay for stockpiling potassium iodide tablets that are used to block the absorption of radioactive iodine by the thyroid gland in the event of a meltdown.

The situation with regard to 65 ex-Soviet reactors was only mentioned in passing, but all present agreed that the old Soviet electrical grid has not been examined for Y2K problems and will be in great danger in the new year. Most of these reactors do not have back up diesel generators in the event of coolant failure. We then moved to the subject of accidental launching of nuclear weapons. The DOD representative admitted that the U.S. policy of “hair-trigger alert” was a vexatious problem when combined with date-related embedded chips and glitches in the early-warning and automated communications systems.

However, while we were confidently reassured by the presidential Y2K committee that the weapons cannot launch themselves, the issue of date-related problems inherent in the U.S. and Russian early-warning systems was glossed over.

It is known that although Russia initially assessed its early-warning, communications and control systems for nuclear weapons, it has only completed the second stage of a five-part Y2K remediation process. Even if the necessary funds were made available on an emergency basis, there is scarcely time left to complete the next three stages, which include fixing errors and testing those performing “committed testing” of the fixes.

And to make a bad situation worse, Russia is now “blind” to U.S. intercontinental ballistic missile launches three hours each day due to serious deterioration in its early-warning satellite systems.

The Pentagon is worried that computer-related mistakes in the early-warning systems of either side could be misinterpreted in an atmosphere of mistrust. There were diplomatic moves earlier this year to establish a joint early-warning center, but Russia lost interest after Kosovo, and although the U.S. is attempting to revive this liaison, it is now almost certainly too late to establish this complex system.

Although the Pentagon has checked millions of computer parts and interfacing systems for its own Y2K compliance, it cannot be sure that a single embedded chip could not wreak terrible damage. In fact, Deputy Secretary of Defense John Hamre said on Oct. 19, 1998, “Probably one out of five days I wake up in a cold sweat thinking [that Y2K] is much bigger than we think, and then the other four days I think maybe we are on top of it.” There are two viable solutions, which if implemented now will avert these possible nuclear-related events.

* Provide alternative electrical generators--solar, turbine, fuel cells or wind driven--at all the nuclear reactors in the world. We also recommend taking all 433 nuclear reactors offline during the period leading up to and immediately after Jan. 1, to decrease the risk of multiple failures.

* Physically take the 4,400 Russian and U.S. strategic weapons off hair-trigger status--de-alert them.

There is only just enough time to implement these steps, but with committed global leadership, four and a half months should be enough.