Have you ever wondered what would happen if satellites orbiting above us began colliding with one another? It’s probably best not to dwell on it, because the consequences could be catastrophic. This scenario was examined by Sarah Thiele of Princeton University and a team of researchers who analyzed satellite traffic in low Earth orbit (LEO).
Their calculations show that across all megaconstellations, a so-called “close approach” — defined as two satellites passing within less than 1 kilometer of each other — occurs on average every 22 seconds. Within the Starlink constellation alone, such an event happens roughly every 11 minutes, and each satellite performs an average of 41 avoidance maneuvers per year to prevent collisions with other objects.
While this may appear to be a well-functioning system, the authors emphasize so-called “edge cases” — rare but extreme events that most often lead to the failure of complex systems. In the case of LEO, the primary risk factor is solar storms. On one hand, they heat the upper atmosphere, increasing drag and positional uncertainty, which forces satellites to perform more frequent orbital corrections and consume additional fuel. On the other hand, solar storms can damage navigation and communication systems, preventing satellites from maneuvering at all.
The authors cite the so-called Gannon storm of May 2024, during which more than half of all satellites in low Earth orbit had to burn fuel to correct their trajectories. Even more dangerous, however, is the loss-of-control scenario. The study introduces a new risk metric — the CRASH Clock (Collision Realization and Significant Harm) — which estimates how quickly a serious collision could occur after control is lost. According to the calculations, in June 2025 this window was approximately 2.8 days, compared to 121 days in 2018, before the era of megaconstellations.
Particularly alarming is the finding that even a 24-hour interruption in the ability to command satellites corresponds to roughly a 30% risk of a catastrophic collision — one that could trigger the Kessler syndrome. While the full development of this phenomenon — a cascading chain of collisions that renders orbit unusable — unfolds over decades, a single major incident could initiate it almost immediately.
The researchers stress that solar storms offer very little warning time — typically just one or two days — during which there is little that can be done beyond attempting to secure systems. They also point to a historical precedent: the Carrington Event of 1859, the strongest known solar storm. If a similar event were to occur today, it could paralyze satellite control for longer than a few days, potentially leading to a permanent loss of access to low Earth orbit.
Importantly, the authors do not call for abandoning megaconstellations. Instead, they argue for a realistic assessment of risk. In their view, the current model of exploiting LEO delivers enormous technological benefits, but without better coordination mechanisms, system resilience, and crisis management, it could expose humanity to a multigenerational loss of access to space.