A single catastrophic collision between satellites in low Earth orbit could be expected within just 72 hours if every spacecraft operator simultaneously lost the ability to track threats and steer out of the way. That is the central finding of a preprint study posted by Alfonso Gonzalez and colleagues in Cornell University’s Department of Mechanical and Aerospace Engineering, who built a probability model using orbital population data captured on June 25, 2025, when more than 40,000 tracked objects were circling the planet.
The researchers call their metric the CRASH Clock, short for Collision Realization And Significant Harm. It is not a prediction that disaster will strike this week. It is a stress test: a way to measure how quickly a crowded orbit turns dangerous once the safety net of collision-avoidance maneuvers is removed entirely. For the billions of people who rely on GPS navigation, weather satellites, and broadband internet delivered by low-altitude constellations, the thinness of that safety net matters.
“The CRASH Clock is designed to quantify the time to first catastrophic collision under a total loss of conjunction screening and maneuvering capability,” the preprint states, framing the 72-hour figure as a modeled expectation value rather than a live countdown.
How the model works
The CRASH Clock uses Poisson-probability calculations to estimate the likelihood of one or more destructive impacts over time. Its inputs are straightforward: the density of objects at specific orbital altitudes, the relative velocities at which those objects cross paths (often exceeding 7 km/s), and a conjunction threshold of 1 kilometer, the distance at which a tracking error or delayed response could turn a close call into a direct hit. Those inputs are drawn from the June 2025 orbital snapshot, and the full-text version of the paper shows the collision probability climbing steeply as the hours tick by.
The scenario is deliberately extreme. It assumes no spacecraft performs any avoidance maneuver and that operators suffer a complete loss of situational awareness. No major satellite operator has ever publicly experienced that kind of total blackout. But the model is designed to answer a specific question: if everything failed at once, how much time would the world have before physics took over?
The answer, roughly three days, lands in an uncomfortable overlap with real operational timelines. A separate document available through the NASA Technical Reports Server examines how conjunction warnings are handled in practice, including probability-of-collision thresholds that trigger maneuver decisions and a 72-hour planning window for differential drag, a technique in which a satellite adjusts its atmospheric drag profile to nudge its orbit. That document serves as background context for understanding operational response times rather than as a direct validation of the CRASH Clock. Still, the overlap is notable: 72 hours is both the window in which a first collision becomes statistically expected without intervention and the window within which at least one mitigation technique can still work, provided tracking data and command links remain intact.
Why debris cascades are the real fear
A single collision in orbit does not stay a single event for long. The European Space Agency has documented the mechanics in detail: when two objects collide at orbital speeds, the impact generates hundreds or thousands of new fragments, each large enough to destroy another spacecraft. ESA uses the term Kessler Syndrome to describe the self-reinforcing cycle in which debris breeds more debris until entire orbital bands become unusable, a scenario the agency has illustrated in its debris projection visualizations.
This is not purely theoretical. In 2009, an active Iridium communications satellite and a defunct Russian Cosmos spacecraft collided at roughly 790 kilometers altitude, producing more than 2,000 pieces of trackable debris, many of which remain in orbit today. In November 2021, Russia’s deliberate destruction of its own Cosmos 1408 satellite in an anti-satellite weapons test added over 1,500 trackable fragments to an already congested band. Both events forced the International Space Station and other spacecraft to perform emergency avoidance maneuvers in the weeks that followed.
ESA’s 2024 Space Environment Report recorded more than 50 collision-avoidance maneuvers performed by ESA-operated spacecraft alone in a single year, a figure that does not include maneuvers carried out by other operators using ESA-shared conjunction data. Across all operators globally, the frequency of these maneuvers has risen sharply as constellations have grown.
“The congestion problem is not hypothetical,” said Holger Krag, head of ESA’s Space Safety Programme, in a 2024 agency briefing on orbital sustainability. “Every additional object we place in orbit without a reliable disposal plan raises the collision probability for everything else sharing that altitude.”
What the model does not answer
The CRASH Clock’s 72-hour figure is a modeled output of one worst-case scenario, not a countdown running in real time. Several important unknowns sit between the model and reality.
The first is whether a total operational blackout could actually happen. A cyberattack on space surveillance networks, a geopolitical rupture in data-sharing agreements, or a cascading software failure across ground stations could each degrade the system, but no publicly cited incident has approached the level of simultaneous, system-wide paralysis the model assumes. The paper does not attempt to estimate the probability of its own trigger condition.
The second is autonomous onboard collision avoidance. SpaceX, which operates more than 6,000 Starlink satellites as of early 2025, has built an autonomous collision-avoidance system that uses onboard processing and propulsion to execute maneuvers without waiting for ground commands. How well that system, or similar ones on other constellations, would perform during a prolonged loss of ground-based tracking data has not been publicly quantified. If autonomous systems could keep functioning independently, the effective safe window could extend well beyond 72 hours.
The third is the current state of the orbital population itself. The CRASH Clock’s probabilities are tied to the June 2025 census of tracked objects. As of May 2026, no updated official catalog has been incorporated into the model. If launches have continued to outpace deorbiting and debris-removal efforts, the real risk could be higher than the study suggests. If operators have successfully deorbited aging satellites and kept debris growth in check at key altitudes, the margin could be wider.
Finally, international data-sharing arrangements for conjunction warnings remain only partially transparent. Multiple national and commercial tracking networks contribute to the global picture, but there is no single, publicly audited benchmark for how resilient those arrangements would be under stress, or how quickly alternative networks could compensate if a major provider went offline.
Moriba Jah, an astrodynamicist at the University of Texas at Austin who studies orbital debris and space traffic management, has noted in public commentary that models like the CRASH Clock are valuable precisely because they force a conversation about systemic fragility. “We do not have a globally coordinated space traffic management system,” Jah has said in interviews. “We have a patchwork, and patchworks have gaps.”
What the 72-hour vulnerability window means for orbital governance
The CRASH Clock’s value is not as a forecast. It is a margin-of-safety measurement, one that tells policymakers and satellite operators how thin the buffer is between normal operations and a runaway collision cascade. Used that way, the model highlights where investment could buy the most time: more resilient tracking infrastructure, standardized protocols for cross-border data sharing during crises, wider deployment of autonomous onboard avoidance, and debris-mitigation rules that prevent orbital densities from climbing unchecked.
For defense planners, the implications are strategic. Low Earth orbit is now critical infrastructure, carrying military communications, reconnaissance, and missile-warning systems alongside commercial broadband and Earth-observation satellites. A 72-hour vulnerability window in a contested environment is a planning problem that existing governance frameworks, many of which were designed when orbit was far less crowded, have not fully addressed.
For insurers and commercial operators, the number reframes risk. Satellite insurance underwriters already price policies based on launch failure and on-orbit anomaly rates. A peer-reviewed validation of the CRASH Clock, which has not yet occurred as of May 2026, could force a reassessment of how systemic risk in orbit is modeled and priced, particularly for constellations that depend on continuous ground contact for safe operations.
The message is not that catastrophe is three days away. It is that the distance between routine operations and irreversible damage may be far shorter than the frameworks governing space traffic currently assume, and that distance is shrinking as orbit gets more crowded.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
