Out in the main asteroid belt between Mars and Jupiter orbits a rock called 6478 Gault. It's about 4–9 kilometers across (it can be hard to get the exact sizes of asteroids), which is pretty typical. It's orbit is a little unusual in that its more elliptical and tipped than for most asteroids, but that's no big deal.
What is a big deal is that it appears to be tearing itself apart.
Observations of Gault taken on January 5, 2019 by the ATLAS telescopes in Hawaii showed that, of all things, it had sprouted a tail! A long one, too, stretching at least 400,000 kilometers — more than the distance of the Moon from the Earth.
But when did this happen? How long before those observations did the tail form? Astronomers went back to look at older data from ATLAS and Pan-STARRS (a large telescope that takes surveys of large swaths of the sky) and found that the tail could be seen on Gault going back to early December. They looked back even further and saw no indication of a tail in observations from 2010 to as recently as January 2018 (after that the asteroid was too close to the Sun in the sky to observe until late in the year).
Asteroids sometimes form tails after being impacted by another, smaller asteroid, so Hubble was called into action to see if the tail showed any signs of forming that way. To their surprise, astronomers saw a second tail coming off Gault!
The second tail was shorter, narrower, and more well defined than the first, indicating it was younger. That's because pressure from sunlight tends to blow on small particles hard enough to disperse them rapidly, making the tail fuzzier.
Using models of how dust moves in these situations, the astronomers estimated the first tail formed on or around October 26, 2018, and the second on December 30. That makes an impact scenario unlikely. Getting hit hard enough to eject a dust cloud is pretty rare, so having a second one within weeks of the first impact is a bet no astronomer would cover.
Could this be a comet in disguise? Maybe a really old one that doesn't have much ice in it, but somehow suffered some event that made some of that remaining ice suddenly warm up, turn into a gas, and eject dust? Perhaps it used to be farther out in the solar system, and Jupiter's gravity recently nudged it into its current path.
The astronomers ran simulations on the orbit of Gault, and found that it's stable even when they projected it 100 million years into the future. That makes it pretty unlikely to be a recent interloper; we've seen lots of comets affected by Jupiter, and they don't last anywhere near this long. So it's much more likely it's been in its current orbit for a long, long time.
So what's going on then?
The Yarkovsky, O'Keefe, Radzievskii, and Paddack effect, that is. I've written about this before, so let me hand it over to past me:
A very odd but important concept in physics is that even though light doesn't have mass, it has momentum. When sunlight hits an asteroid, the asteroid warms up and re-radiates that heat away as infrared light. If the asteroid were perfectly spherical and non-rotating, the only effect would be to push the asteroid farther from the Sun (think of it like a very gentle breeze blowing on the asteroid, giving it a teeny tiny push).
But if the asteroid is spinning, or has an irregular surface (which, of course, they do; they have craters and pits and cracks and so forth), then that light it emits can be sent away at an angle, and this generates a small torque, causing the asteroid to spin faster. Over time, it can significantly change the asteroid's rotation! That depends on many things, like the exact shape of the asteroid, how reflective it is, how close it gets to the Sun, and more. This process is called the YORP effect, due to it being dreamed up by scientists named Yarkovsky, O'Keefe, Radzievskii, and Paddack.
So sunlight can spin up an asteroid, making it rotate faster. That's no big deal in the short term, but in the long run it spells disaster: At some point the rock is spinning so rapidly that the centrifugal force outward on its surface balances the gravitational force inward. If you're a rock sitting on the surface, over time as the asteroid spins faster you feel less and less gravity. You weigh less!
This changes the stresses and pressure of rock piles on the surface. At some point something has to give, and you get a small landslide. Boulders, rocks, pebbles, and smaller dust grains tumble down under the feeble gravity. The smallest grains of dust gain the most energy as things move and jostle, and they wind up getting the most velocity… which, on a small asteroid, can easily exceed the escape velocity. That dust literally leaves the surface, creating a cloud around the asteroid. The pressure from sunlight then blows it away, creating a tail.
Is this what's happening to Gault? Observations from the ground indicate it has a rotation rate of about 2 hours, and it turns out that's almost exactly where you expect the rotational speed to start causing effects like this! Also, the dust is leaving the asteroid relatively slowly, at speeds of under a meter per second (less than walking speed). That's also about what you'd expect from dust launched into space by a landside (and the rotational speed of the asteroid on the surface near the equator is about 2 meters per second, which also gives the dust a kick).
Incidentally, the amount of material in the tails isn't much: 7 million tons in the first tail, and 40,000 tons in the second. I know that sounds like a lot, but those are equivalent to spheres just 82 and 14 meters across; not much compared to an asteroid several kilometers wide.
It all adds up: Gault has been getting spun up by the ethereal breeze of light from the Sun, and is now very close to the point where it'll fly itself apart. We don't know how long that will take — it depends on lots of factors hat are hard or impossible to measure, like how dense it is, how big it is, what the surface is like, and even how rapidly its rotation is speeding up (which, to be fair, isn't impossible to measure, just very hard).
It's estimated that about one asteroid out in the main belt blows itself apart every year! While we don't know when Gault will self-destruct, we can be pretty sure that it will inevitably join those legions of now-debris-fields.
Depending on what stage they're at, these asteroids aren't all that easy to see. Small ones tend to go first, and they can be incredibly faint, so even if they disperse and create a dust cloud, they only get a little bit brighter. That may be why we don't see many… but we do see them. P/2013 R3 fell apart in 2013, for example. As telescopes get better and we scan more of the sky more deeply, you can bet we'll see more.
In the meantime you can be sure astronomers will keep their eyes on Gault. We may be able to watch an asteroid destroy itself in real time, and that's something that's too cool to miss.