Much about the outer solar system is still a mystery.
That's not a huge surprise, because it's very far away. Neptune, the outermost large planet, is about 4.5 billion kilometers out from the Sun, fully 30 times farther out than the Earth. And that's the inner edge of the outer solar system, so we're talking distances that are truly vast. Even big objects out there are hard to see.
It had been predicted for decades that largish icy/rocky bodies would be out there past Neptune, with Pluto just being the biggest of them. Pluto was discovered in 1930, but the next Kuiper Belt Object (KBO), as they're called, wasn't found until 1992. Now we know of over 1,000, but even the biggest are barely more than dots in our largest telescopes. That makes them hard to study.
… but not impossible. In fact, we do know many interesting things about them. For example, many have satellites. The six largest KBOs all have satellites, almost all of which are much smaller than the parent body. It's also known that many smaller KBOs are binaries, pairs of roughly equal-sized objects. Likely these smaller binaries formed as pairs, or formed as separate objects that wound up orbiting each through subtle physical processes early on in the solar system's history.
But what of the big ones? The large size ratios could be due to either the parent body capturing the smaller one, or because the moon(s) formed after a collision ejected debris into space around the parent. Distinguishing these processes isn't easy, but one way is to look at how much light the objects reflect. A small KBO that forms on its own tends to reflect something like 4-8% of the sunlight that falls on it, but one that forms after a collision tends to have a lot more water ice on its surface, so reflects much more light, like 50-100%.
The problem is, to know how reflective an object is, you need to know how big it is — for a given brightness we measure from Earth, a bigger object will have a lower reflectivity (what we call albedo), but if it's smaller it'll be shinier to give us the same brightness. But how do you measure the size of an object that is some tens of billion of kilometers away, and far too small to be resolved?
It turns out you do this by changing your wavelength. Even at frigid outer solar system temperatures, these objects are still above absolute zero, and the laws of physics then demand they emit light due to their "warmth." Just like something at high temperature glows in visible light, something at a lower temperature glows, but at much longer wavelengths.
The Atacama Large Millimeter/submillimeter Array (ALMA) observes at just these wavelengths, and also has such keen eyesight it can separate the moons of some larger KBOs from their parents. Astronomers (full disclosure: One of them is my friend Mike Brown) used ALMA to observe the KBOs Eris and Orcus, to see if they could see their moons Dysnomia and Vanth. Measuring their brightness in these wavelengths, plus knowing their temperatures, leads to an unambiguous determination of their sizes.
They were able to easily separate Orcus and Vanth, and found diameters for them of 910 +- 50/40 km and 475±75. This gives Vanth an albedo of 8%, which is normal for a KBO of that size. Orcus, which is much larger, has an albedo closer to 25%, again about what we usually see from larger objects.
We don’t know what the density of Vanth is, so its mass isn’t clear. A plausible range of densities gives a mass ratio of Orcus/Vanth of 5 to 20. For comparison, the Earth is 80 times the mass of our Moon.
Eris and Dysnomia are a more difficult case, because Dysnomia is faint and difficult to see with ALMA. In fact, they didn't see it clearly in any of their observations! However, combining all their observations (by shifting them to the predicted position of Dysnomia so that the observations can be added together, boosting the signal) does seem to reveal a marginal detection of it. It is known from previous observations that Eris is 2,326 km wide, and using that together with the ALMA observations find a diameter for Dysnomia of 700 km, with a rather large uncertainty of ±115 km. Using that they get an albedo of 4%; low, but again within the range you’d expect for a normal KBO that size.
Both observations strongly imply that these moons are captured objects, and not from reassembled collisional debris.
Except… the relatively low mass ratio between Orcus and Vanth is odd. Models don't indicate something like this system would form together, and the odds of a large KBO capturing another largish one are low. It's possible Vanth formed from a giant grazing collision; that would explain a lot about the system, and we know that happens (our own Moon may have formed this way, and Charon, Pluto's large satellite may have as well). The problem is you'd expect several smaller moons to form from this event; Pluto has four smaller icy moons. None is seen down to a size of 10–20 km, and that's smaller than some of Pluto's moons. So that's weird.
Eris and Dysnomia are even weirder; the moon is way darker than the parent. And their estimate for the size is pretty big, much larger than previously thought. Again, a capture scenario seems unlikely, and Dysnomia is way too dark to be made of icy debris from a collision. It's possible that their calculations are in error since the moon wasn't seen well in the observations, and that's throwing things off. It's also possible (I’d say pretty likely) we just don't know enough about how these kinds of moons form around big KBOs.
But you know what that means: We need more observations! That's always true, but in this case these first ALMA observations show that things out there are pretty weird. I'll note that Orcus and Vanth were 7.5 billion km away when the observations were made, and Eris/Dysnomia were a staggering 15 billion km out! Mind you, Eris is the second biggest known object in the Kuiper Belt (Pluto is slightly larger), yet it's smaller than our Moon and nearly 40,000 times farther away! It's kind of amazing we can figure out anything at all about these objects at that distance.
But that's what astronomers do. We’ve only been spying on these distant frigid words for 25 years or so, yet we've still learned quite a bit. Give us more time. There are plenty of surprises yet to come. I can guarantee that.