Kessler syndrome is a chain reaction in which debris from one collision causes more collisions, so the junk in an orbit keeps growing even if launches stop.
In 1978 two NASA scientists, Donald Kessler and Burton Cour-Palais, published a paper on how often satellites would hit each other. They predicted that fragments from random collisions in low Earth orbit would become an important source of small debris from about the year 2000. Soon after, John Gabbard of NORAD gave the idea a name: the Kessler syndrome.
The idea is simple. Two objects collide and break into many pieces. Some of those pieces are big enough to smash other satellites. Each new collision makes more pieces, which cause more collisions.
A collision needs a pair of objects. Kessler and his colleagues explain that the collision rate rises with the square of the number of objects in orbit, while other sources of debris rise only in step with it. ESA puts it plainly: double the number of objects and the collision risk becomes about four times larger.
Orbital speeds make every hit violent. When the working Iridium 33 struck the dead Kosmos 2251 on 10 February 2009 at 776 km, they met at 11.7 km/s. Both were destroyed, and more than 2,300 trackable fragments were created.
It is not a sudden disaster like in films. Kessler's team stress that the cascade is slow and takes decades at today's densities. But their models found that some bands, around 900–1,000 km, are already above a critical density: there, drag cannot remove fragments as fast as collisions make them. ESA notes that debris is most crowded at 800–1,000 km and near 1,400 km, and that objects above about 800 km generally stay in orbit for many decades.
India has its own example of how long fragments last. A PSLV-C3 upper stage broke up in 2001 and made 371 pieces of debris. ISRO counted 41 of them still in orbit at the end of 2024 — more than two decades later.
Dodging collisions helps, but Kessler's review concludes that new satellites and rocket stages must not be left in orbit after their working life, and that some objects already there will probably have to be removed. India's Debris Free Space Missions goal aims for every Indian mission, government or private, to leave no debris by 2030.
Every satellite in the list shares space with the fragments behind Kessler's warning; those orbiting near 800–1,000 km fly through the most crowded band. Browse tracked satellites
Why the square? If N objects share an altitude band, the number of possible pairs is N(N − 1)/2, close to N²/2 for large N. Each pair has a small chance of meeting, so the collision rate goes as N². Explosions and dead satellites add debris roughly in proportion to N. As N grows, collisions must eventually dominate.
Kessler's 2010 review fits the observed collisions with C = K t³, where C is the cumulative number of collisions, t is time and K is a constant. The cube follows from the square: if the population grows linearly, N = a t, the collision rate is about b a² t², and integrating over time gives (b a²/3) t³.
How big a fragment is dangerous? Ground and on-orbit tests put the catastrophic break-up threshold at about 40 joules per gram of target mass. Check it at 10 km/s: one gram of projectile carries ½ × 0.001 kg × (10,000 m/s)² = 50,000 J. Dividing by 40 J/g gives 1,250 g of target per gram of projectile — the 1 : 1,250 mass ratio quoted in the paper. So a 1 kg fragment can shatter a 1,250 kg satellite. The same tests found that 90 to 100 of the fragments made are large enough to break up another target of the same size.
Stability also depends on decay. The paper notes that a fragment's decay can be predicted to within 10–20% if its area-to-mass ratio and starting orbit are known. Where drag removes fragments faster than collisions create them, a band is stable. Where it cannot — as the models show for 900–1,000 km — the population keeps growing even with no new launches.
NASA scientists Donald Kessler and Burton Cour-Palais, in a 1978 paper. John Gabbard of NORAD later gave the effect its name.
A collision needs a pair of objects, and the number of pairs grows with the square of the number of objects. Doubling the population makes collisions about four times more likely.
A PSLV-C3 upper stage broke up in 2001 into 371 pieces; 41 were still in orbit at the end of 2024.