In 1998, The NASA Near-Earth Object Program Office was established at JPL to coordinate NASA-sponsored efforts to detect, track and characterize potentially hazardous asteroids and comets that could approach the Earth. The primary computational activities of this Office involve the continuous (and largely automatic) use of new data to update the orbits of Near-Earth objects so that their future orbital paths can be examined for close Earth approaches. If the possibility of a future close Earth approach arises, impact probabilities are computed and the circumstances of the Earth approach are noted on the SENTRY (Impact Risk) region of the Near-Earth Object Web site (http://neo.jpl.nasa.gov). When the possibility of a particularly close Earth approach is identified, JPL's computations are compared and verified with a parallel, but independent, effort in Pisa Italy called the Near-Earth Objects Dynamic Site, (NEODyS). Once verified, these events are immediately posted on the SENTRY page. At any given time, there will be several dozen Near-Earth objects on the SENTRY Web site for which a future Earth impact cannot yet be ruled out. But with additional data to improve our calculation of their orbits, the vast majority of Near-Earth Objects will be removed from SENTRY. At the same time, newly discovered objects, with orbits that require more observations, will be added to the SENTRY list so there will always be objects on the Risk Page.
What do we do if we discover an asteroid that may hit the Earth?
Although Hollywood has created some colorful methods for stopping an object that is on a collision path with Earth, no government agency, national or international, has been tasked or accepted the responsibility to stop such an asteroid, should one be discovered. But there have been a number of academic and some technical studies, not to mention numerous movies, on how a devastating asteroid impact might be avoided. Since asteroids outnumber comets 100 to 1 in the inner solar system, the asteroids, rather than comets, represent the majority of the nearer-term threat to our planet.
Because of the wide range of possible sizes, trajectories and warning times for Earth-threatening asteroids, there will be a corresponding wide range in the levels of challenge in providing an appropriate response. Unless there are a few decades of warning time, hazardous asteroids larger than a few hundred meters in diameter will require enormous energies to deflect or fragment. In the rare case of a large threatening asteroid, nuclear explosions that could push or fragment the object might provide a sufficient response.
For the far more numerous asteroids that are smaller than a few hundred meters in diameter, if we have adequate early warning of several years to a decade, a weighted robotic spacecraft could be targeted to collide with the object, thereby modifying its velocity to nudge the trajectory just enough that the Earth impact would be avoided. The spacecraft navigation technology for impacting a small body was successfully demonstrated when the Deep Impact spacecraft purposely rammed comet Tempel 1 on July 4, 2005, to scientifically examine its composition.
Nuclear explosions and spacecraft impacts are two of the more relatively mature options for deflecting Earth-threatening objects and they have been studied in some detail (for example, see Ref. 1). Another option has been suggested for the small subset of asteroids that might also pass close to the Earth a few years prior to the predicted Earth impact. For these unique cases, the pre-impact close encounter affects the asteroid's motion so strongly that a relatively tiny change in its velocity prior to the close approach will be multiplied several fold during the flyby, thus allowing the asteroid to miss the Earth on the next pass. In these relatively infrequent cases, even the very modest gravitational attraction between the asteroid and a nearby "micro-thrusting" spacecraft (nicknamed a "gravity tractor") could provide enough of a change in the asteroid's velocity that an Earth collision could be avoided (see Ref. 2).
Successful mitigation requires that a threatening asteroid must be discovered and physically characterized soon enough to allow the appropriate response; the current NASA Near-Earth Object Observations program is operated with this in mind. But, since the number of near-Earth asteroids increases as their sizes decrease, we are most likely to be hit by the relatively small objects that are most difficult to find ahead of time. As a result, consideration must also be given to the notification and evacuation of those regions on Earth that would be affected by the imminent collision of a small, recently-discovered impactor. However, if the object could be found far enough ahead of time and our space technology used to deflect it from the Earth threatening trajectory, it would be a tremendous demonstration of our space-faring capabilities!
Ref. 1. http://neo.jpl.nasa.gov/neo/report2007.html
Ref. 2. http://neo.jpl.nasa.gov/neo/pdc_paper.html