Space Junk Explained: The Growing Problem in Earth's Orbit

In 1983, a fleck of paint struck the windshield of Space Shuttle Challenger. The fragment was smaller than a millimetre across. It gouged a pit in the glass deep enough that the window had to be replaced.

That single detail explains the whole problem better than any statistic. In orbit, size is almost irrelevant. Speed is what matters, and everything up there is moving very fast indeed.

Why Small Things Hit So Hard

An object in low Earth orbit travels at roughly 7.8 km/s — about 28,000 km/h. Two objects in different orbits can meet at a combined closing speed considerably higher than that, in some geometries approaching 15 km/s.

Kinetic energy scales with the square of velocity, which means impact energy at these speeds is extraordinary relative to mass. A 1 cm aluminium sphere striking a spacecraft at orbital closing speed delivers energy broadly comparable to a small car hitting a wall at motorway speed. A fleck of paint behaves like a bullet. A lost bolt is a catastrophe.

This is why debris cannot simply be shrugged off as litter. There is no practical shielding against a 10 cm object. Spacecraft carry shielding effective against the very smallest particles, and operators manoeuvre to avoid anything large enough to track — but there is a dangerous middle band, roughly 1 to 10 cm, that is too small to track reliably and too large to shield against.

How Much Is Up There

These figures shift as tracking improves and as new events add fragments, so treat them as orders of magnitude rather than precise counts. The important point is the shape of the distribution: the objects we can see and dodge are vastly outnumbered by the ones we cannot.

Kessler Syndrome

In 1978, NASA scientist Donald Kessler published a paper describing a scenario that has haunted orbital planning ever since. His argument was straightforward. As the density of objects in orbit rises, the probability of collisions rises with it. Each collision produces fragments. Those fragments raise the density further, making subsequent collisions more likely.

Past a certain threshold, the process becomes self-sustaining: collisions generate debris faster than atmospheric drag removes it, and the debris population grows even if humanity never launches anything again. Particular orbital shells could become effectively unusable for generations.

This is commonly framed as a sudden catastrophe, which is misleading. Kessler's own modelling described something slower and grimmer — a gradual, compounding degradation across decades. Some researchers argue certain altitude bands may already be past the tipping point, with the debris population there set to grow regardless of future launch behaviour.

The Events That Made It Worse

Three incidents account for a substantial share of the tracked debris population, and two of them were deliberate.

2007 — China's anti-satellite test. China destroyed its own defunct Fengyun-1C weather satellite with a ground-launched missile. The intercept occurred at around 865 km, an altitude where atmospheric drag is negligible and debris persists for centuries. It generated more than 3,000 trackable fragments in one of the worst single debris-generating events in history.

2009 — Iridium 33 and Cosmos 2251. An operational American communications satellite collided with a derelict Russian military satellite at roughly 789 km. This was the first major accidental collision between two intact satellites, and it produced roughly 2,000 trackable fragments. Nobody had manoeuvred, because nobody had appreciated the risk in time.

2021 — Russia's anti-satellite test. Russia destroyed the defunct Cosmos 1408 satellite, generating well over 1,500 trackable fragments and forcing the ISS crew to shelter in their return vehicles as the debris cloud passed. The test drew broad international condemnation and prompted several nations to commit to halting destructive ASAT testing.

Who Is Watching

The largest catalogue is maintained by the United States Space Force through its space surveillance network of radars and optical telescopes, which publishes conjunction warnings to satellite operators worldwide. Commercial trackers such as LeoLabs operate their own radar networks and offer higher-resolution collision-risk data. The European Space Agency's Space Debris Office publishes regular environment assessments and is among the more candid public sources on how the situation is trending.

The awkward truth is that tracking capability has not kept pace with the population it is meant to track. Predicting where a small object will be days from now requires modelling atmospheric drag, solar activity, and the object's tumbling orientation — none of which is precisely known. Most conjunction warnings turn out to be false alarms, and operators must decide whether to spend fuel dodging a collision that probably would not have happened.

What Is Being Done

The first line of defence is not creating debris in the first place. International guidelines have long asked operators to deorbit satellites within 25 years of mission end. In 2022 the US Federal Communications Commission adopted a considerably stricter five-year rule for satellites licensed to operate in low Earth orbit — a meaningful tightening, though it applies only to US-licensed operators.

Modern satellites are increasingly designed to deorbit themselves, either by reserving propellant for a disposal burn or by orbiting low enough that atmospheric drag brings them down naturally within a few years. Starlink satellites operate low enough that a failed satellite reenters within roughly five years without intervention — a genuinely responsible design choice given the number involved.

Actively removing existing debris is far harder. Capturing an uncontrolled, tumbling object in orbit is a difficult robotics problem, and every proposed method — nets, harpoons, magnetic capture, robotic arms — is expensive per object. Astroscale has flown demonstration missions on magnetic capture and close-approach inspection of real derelict hardware, and ESA has pursued its own removal demonstrations. These are genuine technical progress, but the economics remain unresolved: nobody has an obvious commercial reason to pay for cleaning up someone else's abandoned satellite.

The Megaconstellation Tension

Large constellations complicate this picture in both directions. On one hand, thousands of additional satellites unambiguously increase congestion and the number of conjunction events, and every new operator adds coordination overhead.

On the other hand, modern constellation satellites are typically better behaved than the derelict hardware that causes most trouble. They manoeuvre autonomously, they deorbit reliably, and they occupy low altitudes where drag provides a natural cleanup mechanism. The most dangerous objects in orbit are not new satellites but old rocket upper stages and dead satellites at higher altitudes — massive, uncontrolled, and destined to stay there for centuries.

An Honest Assessment

Space debris is not an imminent catastrophe and it is not a solved problem. It is a slow-moving commons problem with the classic structure: the costs of adding to it are diffuse and deferred, the benefits of launching are immediate and private, and the entity best placed to act is a fragmented international system with no enforcement mechanism.

What has improved genuinely is awareness. Disposal requirements are tightening, satellite designers now treat end-of-life as a design constraint rather than an afterthought, and destructive anti-satellite testing has become diplomatically costly in a way it was not in 2007. What has not been solved is the existing population of derelict objects at high altitudes, which will remain a hazard long after everyone currently working on the problem has retired.