title: "Kinetic Asteroid Deflection: Could It Backfire?" date: "2026-06-01" category: "space" author: "Zhigeng"

Kinetic Asteroid Deflection: Could It Backfire?

I. The Ultimate Force Majeure

If we were to catalog the major crises that human civilization might face, the most quintessentially force-majeure among them would be a (planet-destroying-level) small-body impact with Earth.

Though the probability is exceedingly small, should it happen, it would be a truly total extinction event. The example most frequently cited is the impact roughly 60 million years ago that caused the extinction of the dinosaurs. This is also the most unassailable reason—bar none—why humanity, for the sake of future generations and to "not put all our eggs in one basket," must become a multi-planetary species.

Because if such a crisis simply did not exist, humanity could devote its energy to solving other problems. Protect this Earth well, and live together forever—whether happily, or amid ceaseless conflict, chaos, and crises, yet without the overwhelming risk of total annihilation.

(Of course, humanity currently faces many other major crises; this publication will address them one by one.)

In principle, small celestial bodies that pose an impact risk to Earth are those whose orbits may pass within 50 million kilometers of Earth; most belong to the asteroid belt. Their nature may be relatively solid stony or iron asteroids, or—more likely—rubble-pile asteroids, characterized by irregular shapes, uncertain structures, and mechanical behavior and post-impact motion vectors that are difficult to predict.

They typically orbit the Sun on highly eccentric paths, their orbital extents coming relatively close to Earth's orbit. They are influenced not only by the Sun's gravity but also by gravitational perturbations from other large bodies in the solar system—collectively known as near-Earth objects (NEOs), including both asteroids and comets.

According to the latest data as of May 26, 2026, from the Center for Near Earth Object Studies (CNEOS) at NASA's Jet Propulsion Laboratory (JPL): 11,637 asteroids over 140 meters in diameter with potentially hazardous trajectories have been discovered and had their orbital parameters determined; 878 near-Earth asteroids larger than 1 km in diameter, whose impact would cause a global catastrophe, have been cataloged. The number of undiscovered or uncharacterized asteroids—as well as those under 140 meters—is even larger; I won't elaborate further here.

To prevent small-body impacts on Earth, humanity has developed planetary defense engineering, exploring various defense methods. Kinetic deflection is the only method that has been experimentally implemented so far.

In September 2022, NASA's DART mission successfully deflected the orbit of Dimorphos—the smaller body of a binary asteroid system—shortening its orbital period around the primary asteroid Didymos by 33 minutes, and shortening the binary system's orbital period around the Sun by 0.15 seconds. This was humanity's first deliberate "strike" against a small celestial body, proving that kinetic deflection is a viable path.

But a thought has lingered, unshakeable, in my mind: what if we push it wrong? Could it backfire?

This is not about NASA's DART mission in particular—that was a test mission, and the binary asteroid system chosen must have posed a very low risk of Earth impact. In the future, when a genuine intervention is needed, targeting a body whose orbit is genuinely close to Earth's, the likelihood of pushing it wrong would be greater.

One could say: the probability of backfire, like the probability of a small-body impact on Earth itself, is small but non-zero—and must be guarded against.

An asteroid whose original orbital risk of intersecting Earth was not that high—after human kinetic intervention with imperfect foresight, its momentum deviation is amplified by gravitational perturbations, and it ends up actually striking Earth after several orbital cycles.

II. The Logical Chain of Backfire

"Backfire" is not a straight-line inevitability—it's a low-probability causal chain. Each link's deviation may be amplified by the next, ultimately pointing toward an unanticipated end.

Link One: You Don't Know What You're Hitting

Asteroids are most likely not solid iron lumps, and they are certainly irregularly shaped. More likely, they are rubble piles—a loose agglomeration of rocks, dust, and ice held together by weak gravity. Of course, a rubble pile may also contain a hard core or dense clumps.

What happens when you hit one?

The DART mission delivered a stunning figure: the actual momentum transfer efficiency was 2 to 5 times greater than expected. After the probe struck Dimorphos, the asteroid received a push 2 to 5 times larger than predicted (this finding was published in the March 2023 issue of Nature, volume 616; the core measurement beta ≈ 3.6, with a range of 2.5–5× depending on the model used). Why? Because the impact not only transferred momentum—it also ejected a large amount of debris, and the recoil from that ejecta further accelerated the asteroid's displacement. This effect is called the momentum enhancement factor (Beta).

The Beta value depends on the internal structure of the impacted body—and this structure is unlikely to be known with precision beforehand.

There's some mildly good news: according to Aerospace America, the Colorado startup ThinkOrbital is preparing for on-orbit validation of a new X-ray scanning technology for the space domain, aiming for high-detail imaging of orbital targets at considerable distance. This powerful detection capability will likely first be applied to inspecting artificial orbital objects, but one can hope for its eventual application—far in the future—to the internal structure detection of small celestial bodies.

One can imagine that achieving such a goal would be immensely difficult, let alone coordinating it with a planned kinetic-deflection operation. It can only be held as a hope—a hope that it will truly be useful on the day it's needed.

Back to our question:

Mass unknown, density unknown, internal structure unknown, irregular shape—you ram a spacecraft into it, and you cannot precisely predict how the debris will spray or where the overall body will go.

This is the first building block of backfire.

Link Two: The Solar System Is a Vast Web

After deflection, the asteroid takes a new orbit. What will this new orbit experience?

The Sun's gravity, Jupiter's gravity (Jupiter is the largest gravitational player in the solar system after the Sun; its gravitational perturbation can rip an asteroid's orbit into massive deviations), the gravity of Mars and Venus—even a passing asteroid can alter its trajectory.

This is gravitational perturbation.

In the first place, a small body's orbit under perturbation from large bodies has a degree of inherent uncertainty, and long-term prediction naturally tends toward error amplification. Now we've added a "human deflection" into this chaotic system. What kind of gravitational environment will the new orbit fall into?

If the direction and magnitude of the deflection happen to send the body into a region of stronger gravitational perturbation, then decades later, its orbit may deviate into a completely unpredictable direction.

Link Three: Gravity Keyholes—A More Refined Risk

At this point we must introduce an even riskier link in the backfire chain: gravity keyholes.

Gravity keyholes are extremely tiny, specific regions in space—as small as a few kilometers, even a few hundred meters—at the scale of the solar system, almost negligible.

But if a small celestial body happens to pass through such a region, the planet's gravity acts like a precise hook, locking its new orbit onto a path that will intersect the planet in the future.

In other words: the asteroid might originally have flown by harmlessly. Even after being deflected by human kinetic intervention, it still hasn't hit anything—it has merely entered an orbit that will "precisely impact years later."

This is the most unsettling scenario of backfire.

Some might ask: if keyholes are so tiny, isn't the probability of hitting one absurdly low?

Correct—the probability of any single keyhole is indeed very low.

But the problem is that our planetary defense enterprise is itself confronting low-probability events to begin with. Moreover:

Gravity keyholes are not unique; their sizes depend on the mechanical structure of large planets, their distance from the asteroid's orbit, and resonance relationships. Multiple keyhole regions may exist around each planet; each one is a potential risk point.
Before applying kinetic energy, predictions of the asteroid's post-deflection orbit may themselves contain significant errors. The larger the error, the harder it is to estimate how many keyholes it may cover.

The frightening scenario is that this is not a "one-shot" risk but a compounding of risks—deflection error + perturbation amplification + keyhole hook—three links connected, increasing the probability of backfire.

Link Four: Methods of Kinetic Deflection

To date, the space community has proposed several kinetic intervention methods:

Spacecraft impact: the method already executed by the DART mission.

Upper-stage impact: using a spent rocket upper stage from another launch mission to complete a new mission—turning waste into treasure. The intention of turning waste into treasure aligns with the spirit of sustainable spaceflight and sustainable development, and is naturally worthy of particular affirmation; I myself favor this approach (this touches on design philosophy and deserves a dedicated essay later). But as one can imagine, the design requirements would be extremely high. If a purpose-built spacecraft already struggles to accurately predict post-impact momentum behavior, an auxiliary mission involves coordination with the primary mission and the kinetic intervention window—undoubtedly much harder still.

Rock-against-rock: select a smaller asteroid, tow it over, accelerate it, and ram it into the target. Setting aside the engineering difficulty of towing and accelerating, just imagine two irregularly shaped, internally complex giant rocks colliding at high speed—how likely is it that the momentum will shift as expected, without flying apart?

Standoff nuclear detonation: lob a nuclear bomb to the vicinity of the target asteroid and detonate it, using the resulting momentum—blasting surface material off the asteroid and altering its orbit. Picture the scene: how do you anticipate precise blasting? How do you calculate the momentum trajectory of an asteroid after being hit by a nuclear shockwave?

III. Can Fragmentation Save the Day?

Here's another question: if we shatter the small body into fragments, what about the "shotgun" result?

My view is that fragmentation, in most cases, is actually less frightening. Because if the pieces are small enough, they are more easily incinerated in the atmosphere. A body on the order of hundreds of meters, once shattered into pieces tens of meters across, can be handled by the atmosphere—the atmosphere is humanity's most powerful natural defensive shield.

Of course, this also depends on the asteroid's original size. If a kilometer-class body fragments into multiple hundred-meter pieces, that's not a "shotgun"—it's a "MIRV" (multiple independently targetable reentry vehicle). The atmosphere couldn't handle that either.

So the risk of fragmentation is not a straight line; it has different possibilities: the smaller the original body, the safer fragmentation is; the larger the original body, the more risk fragmentation still carries.

IV. Not Inevitable, But Impossible to Ignore

Frankly, the "backfire" concern is not mainstream in the current planetary defense consensus—few even give it thought.

This indifference has its reasons: first, it's related to expectation bias—when people propose a solution, they subconsciously use their preferred expected outcome to obscure troubling counter-expected outcomes. Second, early warning decades in advance, phased deflection, and multiple rounds of orbital verification can already cover most risk scenarios. And besides, gravity keyholes are so tiny that hitting one really would take bad luck.

But DART gave us a "2-to-5-times" surprise—and that makes it impossible not to worry. Because there are too many uncertainties. In the future, if we truly need to deflect an asteroid that poses a substantial threat to Earth, let's hope the biggest surprise doesn't come at the most critical moment.

Looking back at this backfire logical chain: unknown body structure → deflection error → gravitational perturbation amplification → gravity keyhole lock → precise impact years later. Each link alone has low probability, but strung together, it forms a risk that cannot be ignored.

Of course, I write this concern not to oppose kinetic impact experiments. Experiments, after all, remain one of the few effective research methods available to humanity, and can provide data and empirical support for humanity's future self-rescue. And the impact targets chosen for experiments are presumably low-risk; I trust the space community would not select a truly high-risk asteroid for an experiment that could itself increase risk.

I write this concern only to offer one additional avenue of risk-prevention thinking. I hope that when relevant research is conducted, or when one day a real decision is made to use kinetic deflection to defend against a small-body impact, this perspective can be taken into consideration.

The probability of backfire is very low—but it is not zero. This demands that planetary defense engineering adopt more systematic planning, and that when a real decision to conduct kinetic intervention is made, the maximum possible precautions against adverse outcomes are in place.

Thinking this logical chain through is not crying wolf—it's adding one more layer of insurance for the day humanity truly needs to pull the trigger.