I recently wrote about Arrokoth, a double-lobed icy world orbiting the Sun past Pluto, and what we've learned about how it likely formed. There were some details left out, some steps skipped... but it turns out other research fills in those blanks.
The word asteroid literally means "star-like". That's because until very recently, asteroids looked like stars (points of light) in even the biggest telescopes. Some might be hundreds of kilometers across, but the Main Belt, where most of them live, is hundreds of millions of kilometers away.
But then we started using radar to ping ones that passed close to Earth like a cop pings a speeding car, and using the reflected pulse to determine their shapes... and also built spacecraft that went and visited them, observing them up close, sending back detailed images and other data to scientists on Earth. They went from points of light to being worlds.
And we got a surprise. Instead of roughly spherical lumps, a large fraction of them turned out to be elongated. Bizarrely, a decent percentage is even double-lobed, like a bowling pin shape, or contact binaries, two separate objects gently touching at one spot.
Hypotheses abounded on how these might come to be. One idea is that a peculiar effect on sunlight on asteroids (called the YORP effect) could spin them up, causing them to rotate so rapidly they literally fly apart. The pieces could then reaccumulate to form elongated shapes or contact binaries. While this may be true for many objects (we've seen asteroids fly apart from this effect), it only works if they're close to the Sun. The object MU 69, now named Arrokoth, is clearly double-lobed but orbits the Sun out well past Neptune, in the Kuiper Belt! The YORP effect is negligible there, so something else must be at work as well.
Well, it's possible that an impact could shatter the asteroid, and again the pieces come back together to form these weird configurations. Impacts in the Main Belt happen often, and the average speed of such an event is 6 km/sec, enough to shatter an asteroid if the impacting asteroid is big enough.
But why are they then elongated? What was the physics compelling a big fraction of these bodies to come back together in a shape that's decidedly non-spherical?
A group of planetary scientists used computer models to see if they could figure it out. They started with the idea that most asteroids under the size of about 100 kilometers are "gravitational aggregates" — bodies made up of much smaller, rigid bodies all accumulated together and held that way by their mutual gravity. A less technical term, and perhaps more colorful and expressive, is rubble piles, a mix of fragments of different shapes and sizes as opposed to one giant monolithic structure (Bennu, a 1-km wide asteroid visited by NASA's OSIRIS-REx, is a classic example of such a rubble pile).
Under that assumption, they created computer models to look at what happens if you have a bunch of fragments with different sizes, masses, and spins, and let them try to reaccumulate after some catastrophic event splits apart the parent body (they don't model that actual event, as that would complicate things a lot in their study, and was beyond the scope of what they were trying to show). This is pretty hard to do, since keeping track of every little collision between fragments means things change, like their momentum and spin.
Despite the relative simplicity of their model, what they found is fascinating. They ran the simulations over and again, changing the input parameters (mass, size, spin) each time. Typically, they found it takes 3–5 hours for the fragments to resettle after the catastrophic event, which right away surprised me. I would've thought it would take longer.
But it's the shapes of the final objects that are important. They found that the objects were elongated or even double-lobed about 25% of the time! That's very roughly in tune with what we actually see, so they're on to something here. But why exactly did this happen?
What they found is that what happens to the overall shape depends on what happens to the largest fragment in the simulation. Actual experiments with high-speed collisions show that the largest chunk tends to stay at or near the center of the debris field. They find the same thing in their simulations. But, in some fraction of the cases, the largest fragment can get nudged away from center by slightly smaller pieces hitting it. The movement is slow, about walking speed, but it's important.
That's because the smaller fragments still fall toward the center of mass of the system, even when the largest chunk is off-center. This has the overall result of creating either an elongated asteroid, or in some extreme cases, a double-lobed one, with the largest fragment moved to one end of the system, creating the "head" of the contact binary.
Some of their simulations wind up looking eerily close to known objects. In 2005, the Japanese probe Hayabusa visited the 330-meter long asteroid Itokawa and found it to be highly elongated. In one run, the simulations of reaccumulating fragments pretty much nailed Itokawa's shape.
Remember, they're agnostic about what shatters the asteroid. It could be an impact or it could be YORP or it could be something else no one has thought of yet. Interestingly, they find that the density of the fragments isn't that important, so this may well work for icy objects like comets or Kuiper Belt Objects like Arrokoth. In those cases, though, the strength of the fragments may come into play (ice is more ductile than rock), and these models assume the fragments are rigid.
Mind you, these models are really just a first attempt at this, but they do seem to give the right overall shapes and occurrence percentages, so this is an excellent start. I'll be curious to see if they can improve the models to include things like deformation of the fragments, more fragments, and so on, and maybe even be able to distinguish between different kinds of destructive events that start the process. That would be interesting indeed.
I love all this, to be honest. The closer we examine things, even in our own solar system, the weirder they get. And when we discover something weird, scientists jump all over the data to try to figure it out. Understanding the Universe is something humans love to do (and we have a lot of ways we try), and I wonder if, in some ways, that's one of our defining characteristics. It's a delightful thought.