This paper paints a pretty grim picture of how fragile our situation in Low Earth Orbit has become, arguing that the whole orbital house of cards is one bad day away from collapsing. The authors introduce a new metric called the CRASH Clock which estimates how long it would take for a major collision to happen if we suddenly stopped doing active avoidance maneuvers. Back in 2018 that clock sat at a comfortable 121 days, meaning we had months of buffer if things went wrong. As of June 2025 that number has plummeted to just 2.8 days. That is a terrifyingly thin margin because it implies that if a massive solar storm knocked out communications or tracking for just a weekend we’d be looking at a 30% chance of a catastrophic collision.
The data shows that Starlink is the primary driver here with their 550 km shell now being denser than the debris fields from historical anti-satellite tests. The authors found that in this specific shell you’d expect a close approach of less than one kilometer every 11 minutes. It’s not just theoretical either since Starlink satellites are already performing collision avoidance maneuvers roughly every 1.8 minutes across the whole constellation. The system is basically relying on perfect operation and zero errors to keep functioning, which is exactly why the authors compare it to a house of cards.
What makes this really concerning is that we are likely already past the point of stability in some regions. The paper suggests that altitudes above 600 km and specifically the main Starlink shell are already dense enough to support a runaway chain reaction of collisions. Even if a full Kessler Syndrome event takes decades to play out a single major collision now would be like an Exxon Valdez moment for space, creating a mess that forces everyone to operate in a much riskier environment immediately. The 2.8-day buffer is barely enough time to recover from a major geomagnetic storm, let alone a systemic failure, and the fact that we let the margin for error shrink from four months to three days in just seven years is a massive failure of orbital management.
yogthos in space
An Orbital House of Cards: Frequent Megaconstellation Close Conjunctions
https://arxiv.org/abs/2512.09643This paper paints a pretty grim picture of how fragile our situation in Low Earth Orbit has become, arguing that the whole orbital house of cards is one bad day away from collapsing. The authors introduce a new metric called the CRASH Clock which estimates how long it would take for a major collision to happen if we suddenly stopped doing active avoidance maneuvers. Back in 2018 that clock sat at a comfortable 121 days, meaning we had months of buffer if things went wrong. As of June 2025 that number has plummeted to just 2.8 days. That is a terrifyingly thin margin because it implies that if a massive solar storm knocked out communications or tracking for just a weekend we’d be looking at a 30% chance of a catastrophic collision.
The data shows that Starlink is the primary driver here with their 550 km shell now being denser than the debris fields from historical anti-satellite tests. The authors found that in this specific shell you’d expect a close approach of less than one kilometer every 11 minutes. It’s not just theoretical either since Starlink satellites are already performing collision avoidance maneuvers roughly every 1.8 minutes across the whole constellation. The system is basically relying on perfect operation and zero errors to keep functioning, which is exactly why the authors compare it to a house of cards.
What makes this really concerning is that we are likely already past the point of stability in some regions. The paper suggests that altitudes above 600 km and specifically the main Starlink shell are already dense enough to support a runaway chain reaction of collisions. Even if a full Kessler Syndrome event takes decades to play out a single major collision now would be like an Exxon Valdez moment for space, creating a mess that forces everyone to operate in a much riskier environment immediately. The 2.8-day buffer is barely enough time to recover from a major geomagnetic storm, let alone a systemic failure, and the fact that we let the margin for error shrink from four months to three days in just seven years is a massive failure of orbital management.