Chicken Little Alert

By MICHAEL D. LEMONICK

When Timothy Spahr finally knocked off work on Jan. 13, after more than 10 hours on the job, he figured he was at last done for the night. Spahr's task as an astronomer at the Harvard-Smithsonian Center for Astrophysics, in Cambridge, Mass., is to collect reports of asteroids that might one day pass near Earth. On that Tuesday, he had been processing observations from an automated telescope in New Mexico when he noticed a pinpoint of light that might fit the profile. He calculated the object's orbit and, as usual, posted the information on the Minor Planet Center website for other astronomers to see. Then he went off to dinner with a friend.

What happened next guaranteed that Spahr's workday wasn't nearly over. It also triggered a debate among astronomers about how quickly the public should be informed about dangers from space--and how sure scientists need to be before issuing such warnings. Several times in the past, sky watchers have announced that a rogue asteroid might threaten Earth--triggering the usual banner headlines--only to retract the warning a few days later. But while saying "never mind" is embarrassing, it would be much worse to keep a real danger quiet. And that's why Spahr's drawn-out workday was a prime topic of discussion at the Planetary Defense conference organized by the American Institute of Aeronautics and Astronautics and held last week in Garden Grove, Calif.

While Spahr dined, a German amateur astronomer visited the Minor Planet website, noted the new object, called 2004 AS1, and noticed further that its brightness was expected to increase an almost unbelievable 4,000% in the next day or so--an indication that it was approaching with blistering speed. Then he plotted the orbit Spahr had calculated and realized that the chunk of rock, estimated at the time to be about 100 ft. across, was on a direct collision course with Earth--specifically, somewhere in the northern hemisphere--and only days away. At that size, it would probably explode in the atmosphere a few miles up with the force of a one-megaton H-bomb, enough to wreak havoc on anything directly below.

When the German amateur posted an alert on an asteroid watchers e-mail list, astronomers around the world went into high gear. "By the time I got home at around midnight," says Spahr, "there were five messages waiting on my answering machine." Over the next several hours, he and others raced to try to figure out whether Earth truly was in danger. "All of us were initially very skeptical," says Clark Chapman, an astronomer at the Southwest Research Institute in Boulder, Colo. "We thought it was a mistake or bad data or someone playing a trick."

But when Steve Chesley, at NASA's Jet Propulsion Laboratory, checked Spahr's calculations, he came up with a 1-in-4 probability of a strike. "It was a responsible analysis," says Chapman. "It wasn't mistaken in any obvious way." There was one hitch: the asteroid's projected trajectory was based on only four observations over a one-hour period, hardly enough to be definitive. It would take another look to nail down its path for sure.

Usually a threatening asteroid is spotted years in advance. This time, with just days to spare, astronomers had to get their second look right away. So Chesley did some more calculations to find what's called the keyhole--the tiny region of sky where 2004 AS1 should be if the orbit was correct--and put those coordinates out on the Internet. "It clearly wasn't time to make an announcement," says Chapman, who emphatically denies a BBC report that he was on the verge of telephoning the White House that night. "But if we still didn't know the next morning, I think we would have been obliged to alert people."

Fortunately, the wait was not long. At around 3:30 E.T. that morning, Brian Warner, an amateur astronomer from Colorado Springs, Colo., aimed a telescope at the keyhole and found it was empty. 2004 AS1 wasn't going to hit Earth after all, and probably never will--luckily, since it turns out to be more like 1,600 ft. across. Next time, Spahr won't be depending on a sharp-eyed amateur. "Within two days after the incident," he says, "we had software to check for future impacts automatically."