[♪ INTRO].

The Milky Way is more than two million light-years from the next big galaxy, but that doesn’t mean we’re alone out here. There are actually a few dozen smaller satellite galaxies orbiting ours, keeping us company in the neighborhood.

Over the years, they’ve taught us about star formation and how galaxy clusters behave, so they’ve been really useful to astronomers. Except, there’s a problem: Some satellite galaxies seem to be, well, missing. In 1998, two independent laboratories noticed that there’s a big difference between the number of satellites that simulations predict, and the number we’ve actually seen with telescopes.

It’s called the missing satellite problem, and since we’ve noticed it, no one’s been able to figure out exactly what’s going on yet. Our galaxy has at least 100 billion stars, but its satellites are much tinier. Specifically, they’re dwarf galaxies, or those that typically have less than a billion stars.

We’ve been looking at them for centuries, since several of them, like the Large Magellanic Cloud, can be spotted with the naked eye. But it’s only in the last few decades that we’ve given them serious attention. Among other things, they’re great models for how larger galaxies should behave, which has helped scientists test ideas about everything from stellar activity to chemistry.

Understanding them could also help us figure out how our own galaxy formed. So astronomers have put a lot of effort into estimating how many should be out there. Many of these studies have involved computer simulations.

They’re based on the behavior and amount of dark matter in the Milky Way neighborhood, the matter which we can’t see but helps hold galaxies together. They also take into account what we know about the laws of gravity and the universe’s expansion. These models suggest that there should be hundreds or even thousands of satellites out there, including plenty massive enough to be visible through telescopes, but we’ve only seen a few dozen.

Since 2005, the Sloan Digital Sky Survey has found almost 20, and since 2013, the Dark Energy Survey has spotted about 20 more. With other findings, that brings the total to about 60. And when scientists extrapolate and apply those results to unstudied parts of the sky, they only estimate that there are 100 or so satellites floating around.

That’s still significantly less than the simulations predict. And that might mean something is a bit off with our work so far. It could be that astronomers have been calculating the telescope extrapolations wrong.

So right now, some researchers are working on recalibrating the statistics from the Sloan Digital Sky Survey. On the other hand, it could be that something is wrong with the models. The good news is, this probably doesn’t mean our basic understanding of dark matter is wrong.

That would be a hard truth to swallow, and would require a lot more evidence. Instead, it might just mean some small details need to be tweaked, or that other factors are at play. For example, a few cosmologists have thrown around the idea that dark matter might operate differently on different scales.

And research from 2013, published in the Journal of the Italian Astronomical Society, suggests the solution might have to do with the conditions of the early universe. Thankfully, whatever the solution is, cosmologists are no strangers to reconciliation. They come across problems and exceptions in research all the time, so they’ll keep working on this case, too.

Basically, that’s how science works! And there are already some big projects on the way to help us track down those missing satellites. For one, telescopes are getting more powerful all the time, which will help.

In 2022, the Large Synoptic Survey Telescope, operated by AURA, is scheduled to come online. It will be able to look at ten square degrees of the sky at once; an area almost 50 times the size of the Moon. That means it will be able to look at more of the Milky Way and its surroundings in greater detail than any other telescope to date.

It also helps that computers are getting faster, and cosmologists are already using them to enhance their models and simulations. Some simulations, like some being run at Kavli Institute for Particle Astrophysics, pair dark matter simulations with machine learning. They study hypothetical galaxies to figure out how many dwarf galaxies we’d expect to survive in the real universe.

And other models, like the super high-resolution FIRE simulations, model both dark matter and non-dark matter at the same time to get a more accurate view of how dwarf galaxies form in the first place. These studies all hope to emerge with a more accurate picture of our corner of space, which will hopefully help us figure out what’s going on with all those missing satellites. So one day, we’ll hopefully know all of our neighbors in space, not just the bright noisy ones.

And because of that, we’ll have a better understanding of how our galaxy got to be the way it is, and how we ended up here. Thanks for watching this episode of SciShow Space! Now, even though we’ve been studying the universe for centuries, there are still lots of puzzles we haven’t solved yet, like one about diffuse interstellar bands.

These are mysterious light readings scientists have been picking up for nearly a hundred years, and if you’d like to learn more about what they might mean, you can watch our episode all about it. [♪ OUTRO].