The CRASH Clock paper is now peer-reviewed!
-
This paper took forever to publish. All the co-authors are astronomers, so we initially started with astronomy journals, and they didn't like it. Not astronomy-enough, I guess. So we tried a space journal instead. Here's the article from Acta Astronautica: https://www.sciencedirect.com/science/article/abs/pii/S0094576526004091
And here's the non-paywalled version on the arXiv: https://arxiv.org/pdf/2512.09643
You may remember I wrote a bunch about this months ago when we first put it on the arXiv. Now it's peer-reviewed, and officially DONE.
@sundogplanets Congrats!
-
This paper took forever to publish. All the co-authors are astronomers, so we initially started with astronomy journals, and they didn't like it. Not astronomy-enough, I guess. So we tried a space journal instead. Here's the article from Acta Astronautica: https://www.sciencedirect.com/science/article/abs/pii/S0094576526004091
And here's the non-paywalled version on the arXiv: https://arxiv.org/pdf/2512.09643
You may remember I wrote a bunch about this months ago when we first put it on the arXiv. Now it's peer-reviewed, and officially DONE.
The idea for this paper came from me looking at an orbital density plot, and wanting a more intuitive way to understand these densities. What does 10^-8 satellites per cubic km even mean?!
-
The CRASH Clock paper is now peer-reviewed! So to celebrate, my co-authors and I wrote an explainer article (because how better to celebrate one article than by writing another... oh academia...)
It's now published in The Conversation Canada: https://theconversation.com/a-new-crash-clock-measures-the-chance-of-satellite-collisions-and-its-ticking-down-fast-283481
@sundogplanets Thanks for writing that Sam, it's really well done.
-
The idea for this paper came from me looking at an orbital density plot, and wanting a more intuitive way to understand these densities. What does 10^-8 satellites per cubic km even mean?!
Everything in Low Earth Orbit is travelling at essentially the same speed. So you can take this density plot, and turn it in to a close encounters plot. The heart of the paper is doing that calculation in two different ways, one analytic, and one numerical.
In Starlink's orbit, the densest part of orbit, close encounters closer than 1 km happen every half hour or so. 1km sounds like a large distance, but remember the speeds are 7 km per SECOND. That is scary-close.
-
Everything in Low Earth Orbit is travelling at essentially the same speed. So you can take this density plot, and turn it in to a close encounters plot. The heart of the paper is doing that calculation in two different ways, one analytic, and one numerical.
In Starlink's orbit, the densest part of orbit, close encounters closer than 1 km happen every half hour or so. 1km sounds like a large distance, but remember the speeds are 7 km per SECOND. That is scary-close.
The CRASH Clock (which is an acronym, I apologize deeply) is our metric for how dangerous orbit is. We download real orbits for everything (satellites, rocket bodies, tracked debris) and calculate how long it would be for a collision to occur if no orbital maneuvers happen.
This is a worst-case calculation. What if everyone suddenly lost control due to a solar flare, a bad software update, or someone hacking Starlink? How long do we have to regain control?
-
Everything in Low Earth Orbit is travelling at essentially the same speed. So you can take this density plot, and turn it in to a close encounters plot. The heart of the paper is doing that calculation in two different ways, one analytic, and one numerical.
In Starlink's orbit, the densest part of orbit, close encounters closer than 1 km happen every half hour or so. 1km sounds like a large distance, but remember the speeds are 7 km per SECOND. That is scary-close.
@sundogplanets that’s really interesting
-
The CRASH Clock (which is an acronym, I apologize deeply) is our metric for how dangerous orbit is. We download real orbits for everything (satellites, rocket bodies, tracked debris) and calculate how long it would be for a collision to occur if no orbital maneuvers happen.
This is a worst-case calculation. What if everyone suddenly lost control due to a solar flare, a bad software update, or someone hacking Starlink? How long do we have to regain control?
We originally wrote this paper nearly a year ago. As of last June, the CRASH Clock was at 5.5 days. It's now down to 2.5 days. It was 168 days in 2018, pre-Starlink.
It just keeps dropping as we launch more satellites into orbit.
This metric shows how completely dependent we are on continued perfect operations in orbit. 2/3 of all satellites today are Starlinks, and they performed 300,000 collision-avoidance maneuvers last year. They have done it perfectly so far. How long can that continue?
-
Everything in Low Earth Orbit is travelling at essentially the same speed. So you can take this density plot, and turn it in to a close encounters plot. The heart of the paper is doing that calculation in two different ways, one analytic, and one numerical.
In Starlink's orbit, the densest part of orbit, close encounters closer than 1 km happen every half hour or so. 1km sounds like a large distance, but remember the speeds are 7 km per SECOND. That is scary-close.
@sundogplanets I hope the people who use OUR orbits know what they are doing. If they fuck it up, I'm going to be seriously pissed.
-
We originally wrote this paper nearly a year ago. As of last June, the CRASH Clock was at 5.5 days. It's now down to 2.5 days. It was 168 days in 2018, pre-Starlink.
It just keeps dropping as we launch more satellites into orbit.
This metric shows how completely dependent we are on continued perfect operations in orbit. 2/3 of all satellites today are Starlinks, and they performed 300,000 collision-avoidance maneuvers last year. They have done it perfectly so far. How long can that continue?
The Conversation article above describes a collision scenario more in detail. It takes months to catalogue new debris from collisions, and in the the mean time, additional collisions can happen. That potential collisional runaway is the Kessler Syndrome, the worst-case scenario in orbit (explainer here: https://theconversation.com/too-many-satellites-earths-orbit-is-on-track-for-a-catastrophe-but-we-can-stop-it-275430)
The CRASH Clock is not a countdown to Kessler, but any major collision in orbit, especially in Starlink's super-dense orbit, will be an extremely bad day.
-
We originally wrote this paper nearly a year ago. As of last June, the CRASH Clock was at 5.5 days. It's now down to 2.5 days. It was 168 days in 2018, pre-Starlink.
It just keeps dropping as we launch more satellites into orbit.
This metric shows how completely dependent we are on continued perfect operations in orbit. 2/3 of all satellites today are Starlinks, and they performed 300,000 collision-avoidance maneuvers last year. They have done it perfectly so far. How long can that continue?
It won't be technology that ends their “perfect” record - it will be an MBA who decides it's more profitable to play the odds than to waste fuel adjusting orbits.
