Every day, four to five Starlink satellites fall from orbit and burn up in the atmosphere. Each one leaves behind roughly 30 kilograms of aluminum oxide nanoparticles, invisible but persistent, suspended in the mesosphere and stratosphere for years. Multiply that by thousands of satellites and decades of operation, and the scale of the disposal problem becomes clear: low Earth orbit’s first generation is aging out, and the industry has no choice but to bring it down.

The numbers are stark. SpaceX deorbited 260 Starlink satellites in the first half of 2026, a controlled burn rate driven by the five-year design life of its satellites and the FCC’s new mandate that LEO operators dispose of spacecraft within five years of mission completion. That mandate, adopted in 2022, replaced the old 25-year guideline with a hard deadline meant to reduce collision risk in an increasingly crowded orbital environment. For constellation operators, it means a constant cycle of launches to replace aging hardware and a growing fleet of satellites that must be actively shepherded to their demise.

From Guidelines to Hard Deadlines

The shift from 25 years to five was regulatory acknowledgment that the old standard could not keep pace with mega-constellations. When post-mission disposal was a guideline, satellites could linger in orbit for decades, gradually decaying under atmospheric drag. But with tens of thousands of satellites planned across multiple constellations, that timeline was untenable. The FCC’s five-year rule forces operators to execute controlled, propulsive deorbits while satellites still have fuel and maneuverability, rather than leaving them as uncontrollable debris.

Controlled deorbit is safer but not free. Each satellite must reserve propellant for end-of-life operations, and operators must maintain tracking, command capability, and collision avoidance throughout the descent. SpaceX reports a post-mission disposal reliability rate exceeding 95 percent, with nearly all defunct satellites brought down under active control. That operational discipline comes at a cost: every kilogram of fuel reserved for deorbit is mass that cannot be used for station-keeping, maneuvers, or payload capacity during the satellite’s working life.

The Atmospheric Ledger

What happens during reentry is becoming a subject of scientific scrutiny. A typical 250-kilogram satellite generates around 30 kilograms of aluminum oxide nanoparticles as it ablates in the mesosphere, according to research published in Geophysical Research Letters. These particles are chemically reactive, slow to settle, and may persist in the atmosphere for decades. The concern is ozone depletion: aluminum oxides can catalyze reactions that break down stratospheric ozone, the layer that shields the planet from ultraviolet radiation.

In 2022, satellites reentering the atmosphere deposited an estimated 17 metric tons of aluminum oxides, a 29.5 percent increase over natural sources of aluminum in the upper atmosphere. With constellations expanding and the first wave of retirements accelerating, that figure is rising. By mid-2026, tracking data suggested several hundred Starlink satellites had already reentered for the year, part of a systematic fleet upgrade. At current rates, the industry could be depositing hundreds of metric tons of aluminum oxides annually within a few years—an order of magnitude increase over baseline levels.

The environmental impact remains uncertain. Atmospheric models are still catching up to the scale of mega-constellation reentries, and the long-term effects of sustained aluminum oxide deposition are not well understood. But the trajectory is clear: reentry is no longer an occasional event. It is a continuous industrial process, and the atmosphere is the disposal site.

The Economics of Disposal

Deorbit is not just an environmental question; it is an economic one. For operators, the cost of bringing satellites down is embedded in every aspect of constellation design: propellant budgets, operational lifetimes, replacement cycles, and launch cadence. Starlink’s five-year design life means roughly 20 percent of the fleet must be replaced annually once the constellation reaches steady state. With over 10,000 satellites on orbit as of late 2026, that implies thousands of deorbits and launches per year just to maintain capacity.

The U.S. Space Force’s recent $52 million contract for deorbit-as-a-service—the first of its kind—signals growing interest in outsourcing end-of-life disposal. The contract, awarded to a commercial provider, will deorbit satellites that were not originally designed with propulsive disposal systems. It is a stopgap for legacy hardware, but it also points to a broader question: as constellations age and operators face mounting disposal obligations, will third-party deorbit services become standard practice?

The business case remains uncertain. Astroscale, a leader in active debris removal, has demonstrated rendezvous and capture technologies with its ELSA-d and ADRAS-J missions, and its ELSA-M demonstrator is slated to capture and deorbit a defunct OneWeb satellite in 2026. The company envisions a multi-client servicing model in which a single deorbit vehicle handles multiple satellites, reducing per-unit costs. But building a servicer, launching it, and executing multiple capture-and-deorbit operations is expensive. Operators must weigh the cost of designing satellites with integrated propulsion against paying for deorbit services years down the line.

Policy Levers: Bonds and Incentives

Some researchers argue that financial instruments could improve compliance with disposal rules. A 2026 study in Nature Communications modeled the effect of performance bonds on post-mission disposal rates. Bonds of $200,000 per satellite, the study found, reduced derelict satellite mass by more than one-third and increased overall welfare by up to 27 percent. Bonds above $1 million showed diminishing returns, suggesting a sweet spot where financial pressure is sufficient to change behavior without imposing prohibitive costs.

The World Economic Forum has estimated that inaction on orbital debris could cost the satellite industry between $25.8 billion and $42.3 billion over the next decade, largely from collision-induced outages and lost capacity. Against that backdrop, upfront disposal costs—whether through bonds, dedicated deorbit propulsion, or third-party services—begin to look like insurance. The question is whether operators and regulators will act preemptively or wait until a cascade of collisions forces the issue.

The Replacement Treadmill

Deorbit is inseparable from replacement. Constellations are not static; they are continuously refreshed with newer hardware that offers better performance, more capacity, or improved reliability. SpaceX has been systematically retiring first-generation Starlink satellites even when they remain functional, replacing them with upgraded models. That strategy accelerates technological iteration but also accelerates the deorbit cycle.

The result is a treadmill: to stay competitive, operators must launch faster, upgrade more frequently, and deorbit more satellites. The pace is dictated not just by hardware failure rates but by market pressure and regulatory timelines. The FCC’s five-year rule compresses the window, ensuring that no operator can park a constellation in orbit and walk away. Every satellite launched today carries an expiration date, and the clock is already ticking.

What Comes Next

The deorbit problem will intensify. Amazon’s Project Kuiper has not yet reached full deployment, and other constellations—Telesat Lightspeed, AST SpaceMobile, and others—are still ramping up. When they do, the combined deorbit rate across the industry will climb into the thousands per year. The atmospheric load of aluminum oxides will rise in tandem, and the environmental questions that are theoretical today will become empirical.

Operators have options. They can design satellites with more propellant, longer lifetimes, or alternative materials that produce less harmful byproducts during reentry. They can adopt deorbit-as-a-service models that shift disposal responsibility to specialized providers. They can lobby for regulatory flexibility or accept stricter environmental standards as the price of operating at scale. But they cannot avoid the fundamental problem: what goes up must come down, and the bill for bringing it down is coming due.

Low Earth orbit was once a sparse, forgiving environment where deorbits were rare and atmospheric impacts negligible. That era is over. The new normal is a high-tempo, high-turnover regime in which satellites are disposable, deorbits are routine, and the atmosphere is part of the industrial process. How the industry manages that transition will determine not just the sustainability of LEO constellations, but the broader viability of space as a commercial domain.

Sources