Asteroids really do cross Earth's path
Millions of asteroids orbit the Sun, mostly in the main belt between Mars and Jupiter, but a subset called near-Earth objects (NEOs) have orbits that bring them close to Earth's own path around the Sun. Space agencies track these with telescopes, calculating each orbit precisely enough to know, decades in advance, whether a particular rock has any realistic chance of an impact. As of the mid-2020s, no known asteroid poses a significant impact risk in the next century — but the search continues because smaller, harder-to-spot objects are still being catalogued, and a genuinely dangerous find would need years of lead time to act on.
Impact energy scales with mass and with the square of velocity (kinetic energy = ½mv²), which is why size matters so much: a 20-metre object like the one that exploded over Chelyabinsk, Russia in 2013 released energy equivalent to roughly 400,000–500,000 tonnes of TNT and still shattered windows across a city, while an object a few kilometres wide, like the one linked to the dinosaur extinction 66 million years ago, releases energy on a scale that can alter the global climate for years.
Why you cannot just blow it up
Popular fiction favours detonating a nuclear weapon on the asteroid, but for most realistic scenarios this is the worst option, not the best. An asteroid is typically a loose, weakly-bound "rubble pile" rather than a single solid rock, and a poorly timed or poorly placed explosion can fracture it into several large pieces still on a collision course — trading one big impact for several smaller ones spread over a wider area, potentially still catastrophic. Nuclear deflection is reserved as a last resort for very short warning times or very large objects, and even then the goal is usually to vaporise a thin surface layer to push the object off course, not to shatter it outright.
The two methods that actually work: kinetic impact and gravity tractor
The best-tested real approach is the kinetic impactor: slam a spacecraft into the asteroid at high relative speed to transfer momentum and nudge its velocity by a tiny amount. NASA's DART mission did exactly this in September 2022, striking the small moonlet Dimorphos (orbiting the larger asteroid Didymos) at about 6.1 km/s. The impact shortened Dimorphos's orbital period around Didymos by roughly 32–33 minutes — far more than predicted, because the impact ejected a large plume of debris whose reaction thrust added extra push beyond the direct momentum transfer alone. It was the first time humanity deliberately and measurably changed the trajectory of a celestial body.
momentum transferred ≈ β · m_impactor · v_impact β (beta) momentum enhancement factor from ejected debris (β > 1 typically) m_impactor spacecraft mass at impact v_impact relative closing speed
The other proven-in-concept technique is the gravity tractor: park a spacecraft near the asteroid for months or years and let their mutual gravitational attraction — however tiny — slowly tug the asteroid off its original path. It exerts far less force than a kinetic impact, but it is gentle, precisely controllable, and does not risk fragmenting the object at all, which makes it attractive for smaller nudges planned very far in advance.
Lead time is everything
A deflection strategy's viability depends almost entirely on how early you start. Nudging an asteroid's velocity by even a few millimetres per second, applied a decade before a predicted close approach, compounds over that decade into a course change of thousands of kilometres — easily enough to turn a hit into a miss. The same nudge applied with only months of warning barely moves the impact point at all. This is why the practical planetary-defence strategy is not a single dramatic last-minute mission, but decades of patient telescope surveys to find and track objects while there is still time to act gently.
Frequently asked questions
Has humanity actually deflected an asteroid before?
Yes. NASA's DART mission deliberately struck the asteroid moonlet Dimorphos in September 2022 and measurably shortened its orbital period by roughly half an hour, the first confirmed demonstration of deliberately changing a celestial object's trajectory.
Why not just destroy a dangerous asteroid with a nuclear weapon?
Most near-Earth asteroids are loosely bound rubble piles, and a nuclear detonation risks fracturing one large threat into multiple smaller but still dangerous pieces rather than eliminating the danger. Nuclear options remain under study mainly as a last resort for very short warning times, using a standoff blast to vaporise surface material and push the object rather than shattering it.
How much warning time does a real deflection mission need?
As much as possible — ideally years to decades. A small velocity change applied early compounds over the asteroid's remaining travel time into a large shift in its eventual position, so early detection through sky surveys matters more than the raw power of any single deflection method.
Try it live
Everything above runs in your browser — open Defend Earth and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.
▶ Open Defend Earth simulation