Earth orbit is no longer a vast, forgiving expanse where mistakes unfold slowly. According to a new scientific analysis, it has become a tightly coupled, failure sensitive system where a single breakdown in automation, communication, or situational awareness could trigger a catastrophic satellite collision in a matter of days. Not decades. Not years. Days.
The study introduces a stark new metric for understanding orbital risk called the CRASH Clock. It measures how long Earth’s orbital environment could go without collision avoidance maneuvers before a destructive collision becomes statistically likely. As of mid 2025, that time is just 2.8 days. In 2018, before the rapid expansion of satellite megaconstellations, the same calculation yielded a window of 121 days. The difference is not subtle. It represents a collapse in safety margin so severe that Earth orbit now resembles a system balanced on continuous intervention rather than intrinsic stability.
This is not a theoretical exercise. The CRASH Clock does not model some distant future scenario. It evaluates the current distribution of satellites, debris, and rocket bodies already circling the planet. It assumes no new launches, no geopolitical escalation, and no sudden debris creating events. It asks a single, brutally simple question. If collision avoidance systems stopped working today, how long would it take before a major collision occurred?
The answer is short enough to be unsettling. Within 24 hours of a complete halt in maneuvering, there would be roughly a 30 percent chance of a catastrophic collision between two catalogued objects. Within three days, a collision becomes more likely than not. That collision would not be a minor scrape. At orbital velocities approaching 10 kilometers per second, impacts are violently destructive, producing clouds of high speed debris that threaten other satellites in nearby orbits. The consequences would propagate outward in space and forward in time.
What makes this assessment particularly alarming is that it reframes the space debris problem entirely. For decades, orbital debris has been treated as a slow burn issue. A long term environmental concern where the danger lay in gradual accumulation and eventual runaway cascades unfolding over generations. That risk still exists, but it is no longer the primary threat. The immediate danger now lies in the fragility of the systems we rely on to prevent collisions from happening every day.
Low Earth orbit has quietly transitioned from a regime of passive safety to one of active control. Collisions are no longer rare because the environment is safe. They are rare because thousands of satellites are continuously maneuvering to avoid them. Orbital safety now depends on constant, precise, and error free automation. Any disruption to that process has immediate consequences.
The numbers reveal just how crowded and dynamic the environment has become. Across all of low Earth orbit, close approaches within one kilometer now occur on average every 20 seconds. In the densest orbital shells, particularly those occupied by large commercial constellations, close approaches happen every few minutes. These are not anomalies. They are the expected background condition of the modern orbital environment.
Starlink satellites dominate this risk landscape. Their sheer numbers mean they account for the majority of satellite related close encounters and collision probabilities. In the primary Starlink shell around 550 kilometers altitude, close approaches within one kilometer occur roughly every 11 minutes in that shell alone. Collision avoidance maneuvers are not an occasional corrective action. They are a constant operational necessity.
This reality is already reflected in operational data. In a recent six month period, Starlink satellites executed over 144,000 collision avoidance maneuvers. That averages out to one maneuver somewhere in the constellation roughly every two minutes. Each maneuver alters orbital predictions, introduces uncertainty, and requires coordination with other operators. The system works only because it is continuously managed.
The CRASH Clock strips away that management and asks what the underlying system looks like without it. The result is a sobering measure of systemic fragility. Earth orbit is now a house of cards held upright by automation, tracking accuracy, and communication reliability. Remove any one of those supports, even briefly, and the probability of failure rises sharply.
Importantly, the study does not argue that Earth orbit is already in an irreversible debris cascade. It does not claim that a single collision would instantly render low Earth orbit unusable. What it demonstrates instead is that the margin for error has nearly vanished. The difference between normal operations and a catastrophic event is no longer buffered by time or space. It is buffered by software uptime, data integrity, and human coordination.
This reframing has profound implications because the failure modes are not hypothetical. Software bugs happen. Data outages happen. Ground systems fail. Communication links are disrupted. Space weather interferes with tracking and prediction. Geopolitical events interrupt cooperation. None of these are rare occurrences in complex global systems. What has changed is the environment’s tolerance for them.
Solar storms provide a clear example. During the major geomagnetic storm of May 2024, atmospheric drag increased dramatically, forcing a large fraction of satellites to maneuver simultaneously. Those maneuvers, in turn, increased positional uncertainty across the orbital catalog. For several days, satellite operators were managing not only increased drag but a surge in collision avoidance complexity caused by the very actions taken to maintain safety.
Under those conditions, orbital predictions can degrade by kilometers. Collision probability calculations become unreliable. Maneuvers executed by one operator can force reactive maneuvers by others, creating a cascading management challenge. The CRASH Clock shows that in an environment this dense, even a short period of degraded situational awareness could push the system into a collision regime.
Historical context makes this vulnerability more troubling. The geomagnetic storm of 2024 was strong by modern standards, but it was not unprecedented. In 1859, the Carrington Event produced geomagnetic disturbances at least twice as intense. That event unfolded over multiple days and included successive storm peaks. If a comparable storm occurred today, the impact on orbital operations would be severe. The CRASH Clock suggests that Earth orbit may no longer have the resilience to absorb such a shock without suffering a major collision.
The dependence on error free automation also raises questions about systemic risk beyond natural phenomena. Orbital collision avoidance relies on software pipelines that ingest tracking data, propagate orbits, assess conjunctions, and recommend maneuvers. These pipelines are complex, distributed, and often proprietary. They depend on accurate inputs, consistent updates, and trust between operators.
A software failure does not need to be dramatic to be dangerous. A missed update. A delayed alert. A misconfigured threshold. A temporary loss of tracking data. Any of these could create a window where collision risk spikes without being properly mitigated. In an environment where the expected time to collision without maneuvers is measured in days, even a short lapse matters.
The paper highlights a sobering comparison. In 2018, before the explosive growth of megaconstellations, Earth orbit could have gone roughly four months without collision avoidance before a major collision became likely. Today, that buffer is gone. The system has crossed a threshold where safety is no longer an emergent property of space but an actively enforced condition.
This shift also challenges the way space sustainability has traditionally been discussed. Regulatory frameworks often focus on long term debris growth, post mission disposal timelines, and aggregate object counts. These metrics remain important, but they do not capture the real time operational stress now present in low Earth orbit. A system can be statistically stable over decades while being acutely vulnerable to short term disruptions.
The CRASH Clock provides a way to quantify that vulnerability. It does not depend on speculative future launch rates or assumptions about operator behavior. It is calculated directly from the current orbital population. It can be recalculated over time to track whether conditions are improving or worsening. Most importantly, it translates abstract congestion into time. Time that policymakers, engineers, and the public can understand.
The findings also complicate optimistic narratives about space traffic management solving the problem. Improved coordination and automation are necessary, but they also deepen the system’s reliance on those very tools. Each additional satellite increases not only congestion but also the operational burden required to keep the system stable. The more maneuvers that are required, the more opportunities there are for error, miscommunication, or unintended interaction.
This is not an argument against satellites or their benefits. Modern society depends on space based infrastructure for communication, navigation, weather forecasting, and scientific observation. The issue is not whether satellites are useful. It is whether the current approach to deploying them has quietly eroded the safety margins that once protected orbital space from cascading failure.
The study makes clear that Earth orbit should be understood as a finite, stressed resource. Every satellite added increases density, collision cross section, and operational complexity. Debris and defunct objects compound the problem by occupying space without contributing to active management. In such an environment, sustainability cannot be measured only by long term averages. It must also account for worst case scenarios and system resilience under stress.
Perhaps the most unsettling implication is that the absence of recent major collisions is not evidence of safety. It is evidence of constant intervention. Earth orbit has become a system where failure is prevented not by design robustness but by continuous vigilance. That is a fragile way to operate any critical infrastructure.
The CRASH Clock does not predict the exact moment of the next collision. It does something more important. It reveals how little room for error remains. It shows that Earth orbit is now operating close to the edge of its tolerance, dependent on flawless execution across a web of technical and organizational systems.
In practical terms, this means that discussions about space sustainability can no longer be abstract. They must grapple with the reality that orbital safety is now a real time systems problem. One that can be triggered by software bugs, solar storms, data outages, or geopolitical disruption. One that unfolds on the scale of days, not decades.
Earth orbit has entered an era where stability is no longer passive. It is actively maintained, continuously negotiated, and increasingly brittle. The CRASH Clock does not suggest that disaster is inevitable. It suggests something more uncomfortable. That disaster is now close enough to require everything to keep working all the time.
Source:
This article is based on the December 2025 preprint An Orbital House of Cards: Frequent Megaconstellation Close Conjunctions by Sarah Thiele, Skye R. Heiland, Aaron C. Boley, and Samantha M. Lawler, available on arXiv at https://arxiv.org/abs/2512.09643
