By now, we have all heard about a handful of asteroids that are big enough to level a city or two and have a small but non-negligible chance of hitting Earth. Should we find one heading straight at Earth, what can we do about it, if anything at all?

That is the question addressed by Carusi and colleagues in a study published in the April issue of Icarus, a leading international journal in the planetary sciences. They conducted case studies of two near-Earth asteroids (NEAs) known as 99942 Apophis and 2004 VD17, whose initial orbit estimates indicated measurable probabilities of hitting Earth in 2036 and 2102, respectively. Although refinements to their orbital calculations through intensive follow-up observations have substantially lowered their chances of collisions with Earth, the authors treated the asteroids' initial orbital estimates as full-blown drills to study how such asteroids can be deflected, and to build realistic strategies to prepare ourselves for such events.

The report presents computer simulations that calculate the minimum orbital velocity change we must impart on the asteroids to deflect them away from Earth. A larger velocity change requires a stronger force, and thus imposes a greater technological and financial challenge. To make the exercise realistic, the authors considered performing their deflection maneuvers only when the asteroids cross the orbit of Earth - as the asteroids under consideration are NEAs, they have repeated Earth orbit crossings leading up to the predicted impact dates.

As expected, in general, the authors' calculations show that greater speed changes are needed as the hypothesized impact date comes closer. However, a careful examination also reveals that there are windows of opportunity in which deflection becomes considerably easier largely due to the relative orbital geometry of the asteroids and Earth. For example, in the case of 99942 Apophis, estimated to be a 400 meter chunk of rock, an impactor with 300 kg mass can deflect the asteroid to safety with a carefully angled interception on January 27th, 2020, about 16 years before impact. The authors note that such a deflection maneuver is already achievable with currently existing technologies. However, their study illustrates that things are not always that easy.

The other asteroid they considered, 2004 VD17, has an orbit closely overlapping that of Earth's over a longer span than 99942 Apophis does, and such orbital characteristics makes its deflection much more tricky. Still, the scientists found windows of opportunity such as one in 2021, 81 years before its hypothesized collision with Earth, in which an impactor weighing about a ton could deflect the asteroid away from Earth.

The authors' findings also come with a bit of bad news. While it may be technologically feasible to exert a force large enough to deflect 2004 VD17, their calculations also reveal that the impactor could shatter the asteroid, which is equivalent to converting an approaching rifle bullet into a shotgun round, with consequences that are unpredictable at best. 99942 Apophis, in contrast, should survive the relatively modest forces required to deflect it.

This study by Carusi et al. shows that deflecting real asteroids is within reach of currently existing technologies, given enough time and planning. By definition, NEAs orbit near Earth, so any that threaten us are expected to have a few close encounters with Earth, during which they are easy to find, before the final collision. Therefore, the long planning period considered in this study is realistic.

The current study's strategy will not, however, work well for deflecting objects with highly elliptical orbits such as long period comets; nevertheless, most objects that impose significant threats to Earth are NEAs since their orbits bring them so close to here. The study highlights the importance of efforts such as the SpaceWatch project hosted by the University of Arizona - its goal is to find and track all objects with chances of impacting Earth. It may well turn out that spotting an asteroid heading our way before it is too late is far more difficult than developing technologies to deflect them.