Space debris is the wreckage left behind by seven decades of spaceflight: dead satellites, spent rocket stages, and fragments from collisions and explosions, all still racing around Earth at orbital speed. The European Space Agency (ESA) currently tracks more than 40,000 of these objects. Millions more are too small to track individually but still move fast enough to disable a working spacecraft on impact. And the problem does not resolve itself. Left unchecked, it only grows.
That is why space debris has shifted from a niche engineering issue to an operational concern affecting everything from GPS to internet satellites. Every rocket that reaches orbit leaves behind at least one dead stage unless it is deliberately deorbited. Every collision, whether intentional or accidental, can create hundreds or thousands of new fragments. In that sense, the orbital environment holds a lasting record of everything that has gone wrong in space.
What Counts as Space Debris
Space debris covers anything human-made in orbit that no longer does a job. That includes whole dead satellites that ran out of fuel or failed, upper rocket stages left behind after they delivered a payload, and fragments thrown off by collisions, explosions, and even paint flaking off a spacecraft’s surface over years of sun exposure. ESA’s 2025 Space Environment Report puts the tracked count at over 40,000 objects, a mix of roughly 14,000 active and dead payloads, around 2,000 rocket bodies, and tens of thousands of loose fragments.
Size determines how dangerous a piece of debris is, and it runs opposite to what intuition suggests. A whole dead satellite is trackable and, in principle, avoidable. The real hazard sits in the untracked middle: ESA estimates roughly 1.2 million objects between 1 and 10 centimeters across, too small for current radar and telescope networks to catalog individually, but large enough that a hit at orbital velocity, often 7 to 8 kilometers per second, can punch through a satellite’s structure or a spacesuit. Below that, more than 130 million fragments between 1 millimeter and 1 centimeter add a steady background risk that shielding can partially, but never fully, absorb.
The Kessler Syndrome: Why One Collision Can Cascade
NASA scientist Donald Kessler described the core danger of orbital debris in a 1978 paper, and the mechanism now carries his name. The Kessler syndrome is a chain reaction: a collision between two objects in low Earth orbit (LEO) creates a spray of fragments, those fragments go on to hit other objects, and each new collision produces more debris than the one before it. Past a critical density, the process becomes self-sustaining. New fragments keep arriving even if every launch on Earth stops.
The 2007 Chinese anti-satellite test against the defunct Fengyun-1C weather satellite is the textbook example of how fast a single event can add to the problem. The test used a direct-ascent missile to destroy the satellite outright and created one of the largest debris clouds on record, much of which is still in orbit and tracked today. The anti-satellite weapons explainer covers how that kind of weapon works and why it is the most debris-intensive way to disable a satellite. A cascade does not require a deliberate test, though. The clearest accidental case came in 2009, when a dead Russian military satellite, Kosmos-2251, collided with an active US communications satellite, Iridium 33, at roughly 42,000 kilometers per hour. Both were destroyed, and the collision alone added around 2,000 trackable fragments to LEO.
A cascade can start at any point once density crosses that threshold. Once it does, no single actor can stop it by launching more carefully. The 2025 ESA Space Environment Report found that even a total halt to new launches would not reverse the trend: the debris population would keep growing for more than 200 years on momentum alone, because collisions and breakups now create fragments faster than atmospheric drag pulls the existing ones down. Roughly 10.5 satellite breakups happen per year on average, adding thousands of newly catalogued fragments annually, on top of everything already up there.
How Operators Cope with Debris Today
The near-term response to a growing debris field is not cleanup. It is avoidance, and the volume of avoidance activity is itself a sign of how crowded LEO has become. Every large constellation now runs an automated collision-avoidance system that watches for close approaches and fires thrusters to move a satellite out of the way. SpaceX’s Starlink fleet alone performed roughly 300,000 of these maneuvers in 2025, a 50 percent increase over 2024, driven by the combination of a larger constellation and a denser debris field around it.
Avoidance protects the satellite doing the maneuvering. It does nothing to remove the debris causing the problem, and it does not scale forever. Every maneuver burns propellant that a satellite would otherwise use to stay on station or extend its working life, and every operator running the same math independently means the total maneuver count keeps climbing even as individual satellites get better at dodging. The operational reality that rarely gets stated plainly is this: avoidance and cleanup are solving different problems on different timelines, and improving one does not substitute for the other.
Active Debris Removal: Cleaning Up What’s Already There
A small number of companies now fly missions built specifically to remove existing debris rather than avoid it. The current generation is proving the technique before anyone attempts it at scale. Astroscale, a Japanese-founded company, flew its ELSA-d demonstration in 2021. It used a magnetic docking plate attached to a target beforehand to prove rendezvous and capture. The follow-on ADRAS-J mission went further in 2024, approaching and photographing a real, uncooperative piece of debris, a spent Japanese H-IIA rocket stage, with no capture aid attached to it at all. ADRAS-J completed that inspection phase, then began deorbit operations in 2026. A second mission, ADRAS-J2, is planned to attempt an actual capture and removal of the same object.
ClearSpace, a Swiss company working under an ESA contract, is chasing a similar goal. Its ClearSpace-1 mission uses a four-armed robotic gripper to capture a leftover Vega rocket adapter and pull it down to burn up in the atmosphere. The timeline has already slipped once. The original target object was itself struck by a piece of debris in 2023, a complication that shows how crowded and unpredictable the environment already is.
Every flown mission to date has targeted one specific object, either pre-fitted with a capture aid or studied closely in advance beforehand. Removing debris at the scale needed to meaningfully cut collision risk, thousands of objects rather than one, remains a capability nobody has demonstrated yet. These missions have proven something narrower and still valuable: that rendezvous, close inspection, and controlled capture of real orbital debris are achievable engineering steps. That is the groundwork removal at scale will eventually stand on.
Active debris removal methods also include kinetic capture tools like nets and harpoons. Built by Surrey Satellite Technology Limited for an Airbus consortium, the RemoveDEBRIS mission deployed from the International Space Station in June 2018. The spacecraft captured released target CubeSats using an October 2018 net and a February 2019 harpoon fired at 20 meters per second. As a single-mission cost figure instead of an industry-wide estimate, the European Space Agency awarded ClearSpace a contract worth approximately €86 million in December 2020 to fly ClearSpace-1.
What Would Actually Change the Trajectory
Slowing the debris problem takes two separate efforts working together, and neither alone is enough. The first is prevention. Rules that require operators to deorbit a satellite or rocket stage within a set number of years after its mission ends keep fewer new objects joining the long-term population. The second is removal: missions like ADRAS-J and ClearSpace-1 that go after debris already up there. Prevention rules cannot touch what has already accumulated. ESA’s own modeling backs this up. Prevention alone slows the growth rate. It does not reverse it, because existing debris keeps colliding and fragmenting no matter how carefully new satellites are launched.
Watch the pace of active-debris-removal missions, more than the total object count, which will keep climbing under any realistic near-term scenario regardless of how the removal effort goes. One mission would mark the real turning point: a removal that captures an object nobody prepared in advance, rather than one fitted with a docking plate ahead of time.