Crowding Low Earth Orbit

AFor thirty years the standard account of crowded orbits has run in a single direction: satellites collide, each collision produces fragments, those fragments cause collisions of their own, and eventually a shell of debris closes the region to everybody. Textbooks and briefing papers still open with that same diagram. The scenario is not fanciful, and the arithmetic behind it is sound. It has nevertheless come to occupy more of the public conversation than it now deserves. The constellations launched since 2019 fly low enough that the thin air at their altitude drags a dead spacecraft back down within a few years, so that a fleet effectively sweeps up after itself, provided the individual satellites still answer to commands when the moment comes to switch them off. The hazard has not disappeared; it has changed address. The change of address matters more than it may sound. What follows an orderly disposal is a fireball, and a fireball leaves something behind, which is precisely the part of the story that neither the operators nor their fiercest critics thought worth raising while the argument remained fixed on fragments of metal travelling at eight kilometres per second.

BThe material budget therefore deserves more attention than it gets. A satellite of the current generation weighs a few hundred kilograms, of which roughly a fifth is aluminium, and on re-entry almost all of that metal is converted into oxide particles small enough to remain aloft in the stratosphere for years rather than settling out within days. By 2024 something close to two spacecraft were re-entering every day, and the filings lodged with regulators imply between five and ten times that rate before the decade is out. Nothing comparable has been attempted before. The quantities themselves are not in dispute. No natural process delivers that compound in quantity. The natural flux of meteoritic material is usually produced in reply, and it is genuinely large, but the comparison is misleading in the one respect that matters here: meteoroids deliver mostly silicates and iron, whereas the artificial flux is dominated by a compound that natural processes supply only in trace amounts, and the layer receiving it happens to be the layer in which the ozone that shields the surface is both made and destroyed.

CInes Bergqvist, an atmospheric chemist at the Uppsala Aerosol Laboratory, has done more than anyone else to attach numbers to the question. Sampling at that height is slow and expensive work. Between 2019 and 2023 her group sampled the stratosphere from a high-altitude aircraft on 46 separate flights, and found aluminium in about 10 per cent of the sulphuric acid particles collected above 19 kilometres, a proportion well beyond anything that meteoric input alone can account for. Bergqvist argues that such particles can catalyse the destruction of ozone in much the way that volcanic aerosols are known to do, though she is careful to add that the reaction has so far been demonstrated in a laboratory rather than in the sky itself. Her own caution is worth recording. Her sampling was confined, moreover, to a single band of latitude in the northern hemisphere, and the models that carry her measurements forward assume a rate of re-entry which no operator has yet published in full.

DYusuf Aldana, who studies space policy at the Lagos Institute for Orbital Governance, dismisses the atmospheric alarm as premature and, worse than premature, as a distraction. In his account the decisive weakness of the present arrangement is legal rather than chemical: a constellation is licensed by whichever state its operator files in, disposal rules differ sharply from one state to the next, and an operator that finds a regime demanding may simply file somewhere else. The pattern is easy to overlook. He counted 31 large constellations approved between 2015 and 2023 and found that 12 of them were registered in jurisdictions carrying no binding requirement to remove a satellite from orbit at all. Aldana maintains that no chemical finding, however alarming, will alter behaviour while that gap stays open. Not every loophole is used, of course. His own evidence, it should be said, stops at the filings: he does not show that satellites licensed under the weaker regimes have in fact been abandoned more often, and two of the twelve operators he names have since adopted disposal commitments of their own accord.

EBoth men are describing one failure from opposite ends, and the more useful question is which remedy would survive contact with the industry. Capture vehicles, the technology that attracts the most photographs, remain implausible at scale: retrieving one dead satellite costs roughly what it costs to launch a working one, and a population of thousands cannot be tidied away a single object at a time. A disposal deposit, refunded when a spacecraft is verified to have left orbit, would be cheap to administer and would bite on precisely the operators Aldana identifies, though only if the major launching states adopted it together, since a levy imposed by one of them alone merely moves the paperwork elsewhere. The choice between them is finally political rather than technical. On the atmospheric question this article takes a deliberately cautious line: Bergqvist's measurements are the best available, but a reaction demonstrated in a flask is not yet a reaction proven at 20 kilometres, and the case for restricting launches on those grounds is not yet made. Neither man disputes the other's figures. What can be said is that an industry which disposes of its waste by vaporising it has not closed the account; it has merely postponed the audit.