At present, the most threatening known asteroids all have a relatively low probability of ever striking Earth. One small body called 2007 VK184, which is 130 meters (425 feet) in diameter, has about a one in 3,000 chance of smacking into the planet in 2048. A much larger asteroid known as 1999 RQ36, 560 meters (1,800 feet) across, packs a more significant impact probability of one in 1,400, but not for another 160 years. It's worth noting that over time these odds of impact are often revised, sometimes negating the objects' threat entirely, as further observations better define NEO orbits or as those orbits are deformed by the gravitational pull of celestial bodies.
"That's one of our main problems—as soon as you make a very close approach to a planet, including Earth, the subsequent motion gets much less certain," Yeomans says. "The small errors you have in the orbit get magnified as a result of the close approach."
One tiny asteroid that did strike, burning up on entry, provided a test of CSS and NASA's efforts. In October, a CSS observer picked up a relatively common two-meter (six-foot) object at close range, and astronomers correctly predicted that it would disintegrate over northern Africa the next night. Although last-minute discovery of a larger NEO would be problematic for evacuations, Yeomans notes that being able to pinpoint the location of a brilliant fireball even hours in advance can have great geopolitical value. Armed with accurate NEO information, scientists can warn "countries that might be arguing with one another that this kiloton blast in the sky is not a man-made event," he says. "What if this thing were spotted over the border between Pakistan and India? It could have been a real problem had it not been predicted ahead of time."
For larger inbound NEOs, of course, early detection might allow human intervention to skirt catastrophe. Fulfilling lawmakers' mandate for tracking 90 percent of 140-meter objects would "retire 99 percent of the risk to Earth from all objects of all sizes," Yeomans says. "By retire, I mean, if we know they're coming and they pose a threat, we have the technology to deal with it"—options include detonation or disrupting the object's orbit by ramming into it or tugging it off-course with a nearby spacecraft. "There's any number of ways to mitigate," he says, "but you have to find them well in advance of a threatening encounter in order to undertake any of them."