Eight billion years ago — when the Universe was roughly one-third of its current age — two dwarf galaxies slammed into each other. The stars and dark matter in those galaxies managed to pass by each other without too much trouble. But things weren't so easy for the vast amount of gas contained in the two galaxies. The force of the collision transformed that gas into new galaxies and several brightly shining star clusters that are arranged in a line connecting the two dwarf galaxies.
That scenario explains a shocking observation that researchers made five years ago: two galaxies without any dark matter. The pair of galaxies were such a mystery to astrophysicists because dark matter — an elusive substance that makes up 85 percent of everything in the universe — is an apparently indispensable ingredient in the formation of galaxies.
The new research is presented in an article published Wednesday in the peer-reviewed journal Nature.
IE sat down with astrophysicist Mike Boylan-Kolchin, an associate professor at the University of Texas at Austin who studies galaxy formation, to discuss the new findings.
Interesting Engineering: What is dark matter?
Mike Boylan-Kolchin: Dark matter is just a catch-all name for something that we don't know exactly what it is. We know that it's something that behaves like matter in the sense that it dilutes as the Universe expands, but it doesn't interact with electromagnetic force. So, it doesn't interact with light, other than gravitationally. It interacts very weakly with normal matter, and that means that it can be very hard to detect.
IE: If it's so hard to detect, how do researchers know it exists?
MB: We see evidence for it all over the place, such as in the motions of stars and galaxies. We know that there's a lot more mass in galaxies than you can account for just by adding up all of the visible components of the galaxy. That’s what got people started talking about dark matter in a sort of very serious way.
We also see evidence for it on much larger scales. The Big Bang model requires dark matter for the universe have enough structure to grow from the very earliest phases through to the present day. Without dark matter, we can’t get from the initial conditions to what we see today. Dark matter is essential for getting the large-scale distribution of matter we see around us today.
IE: How did researchers determine that there is so little dark matter in the system they describe in the new paper?
MB: They measured all of the regular atomic matter — it's mostly stars in this kind of system. If you know how much mass there in a system, then you know how fast things should be moving, based on Kepler’s laws. If I know how fast things are moving, that tells me how much mass there is inside those orbits.
If you did the calculation to determine the mass of the Sun based on the velocity of the Earth’s orbit and came up with a figure that was ten or 100 times more massive than the actual mass of the Sun, that might tell you there was extra mass inside of the Earth's orbit. Of course, we don't see this for the Earth moving around the sun, but it’s typically what you see in galaxies. There’s a missing mass that’s attributed to dark matter.
In this galaxy, they didn't see any discrepancy like that. They could basically account for all of the motion of the stars with just the mass that they saw in the galaxy, so there is very little need to invoke any dark matter in this galaxy in particular.
IE: How did astrophysicists react to this surprising observation when it was originally published five years ago?
MB: I think healthy skepticism is the right tag for that. People were intrigued, and they were also rightly trying to find where some errors could have been made. There was a question about how far away this galaxy is. That's important because it affects our calculation of its mass. People were questioning some other aspects, too.
There was a great iterative process where people kept bringing up potential ways this measurement could have been wrong. The authors kept going back and saying, “Well, okay, that's a good idea. We've checked it, and here's why we don't think it's appropriate.”
I think it also spurred a lot of research from other groups. I've been part of some that have tried to say, “Hey, do we see these in our simulations now that we know they might exist? Can we go out and look and see? Should we have expected this had we known to look for it?”
IE: What did your simulations reveal?
MB: We found some galaxies that looked like this one after we went back and looked at them. Now, the simulations weren't meant to reproduce this particular system. It's not like we went to stone tablets and found that Moses had predicted this. It was more that we looked at the best, biggest, and highest-resolution simulations and found that, yes, we can see things like this system developing, based on current assumptions. The simulations can give us clues as to how they might have formed.
IE: Is it fair to say that you were checking to make sure the math worked?
MB: Yes, the simulations let us ask if scenarios like the one presented in this paper are plausible. Do they work in detail, as opposed to just a hypothetical scenario? Does it give you the right distribution of mass? Is the right amount of collision velocity required for this scenario?
IE: Did your colleagues discuss other hypotheses to explain these observations when they came out five years ago?
MB: Other models have been invoked to talk about these observations. A lot of them involve some strong form of interaction between the galaxies in question, and either another galaxy — may be the biggest galaxy in the middle — or something else that would allow the galaxies to be stripped of dark matter and to be puffed up to such a large size. The explanations typically have involved some kind of common denominator of interactions between galaxies.
IE: How do the authors of the new paper explain their observations?
MB: The idea here is that galaxies don't always live by themselves. For example, our own Milky Way has a whole host of smaller satellite galaxies that orbit around it in the same way planets orbit the Sun. There are lots of small galaxies in orbit around the Milky Way.
The scenario in the new paper is that two galaxies were orbiting around a bigger galaxy and smashed into each other. Now, that's pretty rare. Even rarer is that one of them had to come from outside of the system and happened to smash into the galaxy that already lived in that system. That's why they're smacking into each other at a very high velocity.
When they smash into each other, the gas in the galaxies has lots of interactions, so it stays close to the collision point, at least initially. The dark matter doesn't feel anything except gravity, so it just passes through. That separates the dark matter from the regular matter — the gas — in these galaxies. Once it has this collision, the gas gets compressed to a very high density, which is conducive to forming lots of new stars in these two subsystems, that are then separated from the dark matter the galaxies had contained.
IE: Are you satisfied with this explanation?
MB: It seems to be lining everything up really well, but there are a number of things one could test. They've pointed to a couple of places where we could look for clumps of dark matter that have been displaced from these other galaxies. That's a great follow-up test. They also found some additional galaxies that are along this potential collision path. Understanding the properties of those galaxies better would help us understand whether this is the right scenario.
I think one of the strongest predictions that this hypothesis makes is that the ages of the star clusters in these two galaxies would be the same. That's something we could go out and measure. They know when this event should have happened, so the resulting stars should have formed around the same time. I expect further research on this and further tests to confirm these predictions that this model makes.
IE: Does this model require us to change any fundamental understandings that we had before? Or is it completely consistent with existing theory?
MB: One question that will be interesting to answer is how likely is it for two small galaxies like these to collide at this kind of speed. Is this something that we should expect to find a lot of? Or is this a very rare kind of event that’s kind of a one-off in the universe? Those are the kinds of things we can start to check now that we have numbers about this sort of formation scenario.
The other thing the authors emphasize at the end of the paper is that it might be possible to apply these findings to theories of dark matter in order to understand exactly how non-interactive dark matter is. If dark matter can stick to itself — and there are models of dark matter where it does interact a little bit with itself — then the dark matter lumps in these galaxies wouldn’t travel as far once they pass through each other. If there's a little bit of interaction, they would slow down a little bit. Determining where those clumps actually end up might help us understand the properties of dark matter better, which would be really exciting.
IE: Is there dark matter around me right now?
MB: Yes, there is. There's definitely dark matter everywhere. We think it's very low density. I'd say it's maybe [at the density of] a hydrogen atom per cubic centimeter or something. It's at the level where we are very rarely able to see it on Earth. The way people try to look for these dark matter particles is to develop huge detectors underground where they might interact with the normal matter once in a while.
Since it can't really interact any other way other than gravitationally, it can't clump up to the high density that we get for the regular atomic matter. That happens because atomic matter can radiate energy, cool off, and do things like that. Dark matter doesn't have the means to cool itself off. So, even though it's very important on very big scales, it’s completely unimportant on small scales, like Earth, the Moon, the Solar System, and even the nearest stars.
IE: What does the conversation around these findings tell us about astrophysics as a field? Does it expose any rifts or fundamental differences of opinion?
MB: There's a small minority of cosmologists who don’t believe that dark matter is the best explanation for these phenomena. They think the laws of gravity need to be modified. They have been looking at this observation and asking if it’s evidence that their point of view is correct. Does the fact that this galaxy doesn't seem to need dark matter offer evidence for their theory, which is sometimes called modified gravity-modified Newtonian dynamics.
IE: Would it be fair to call these researchers dark matter skeptics?
MB: Yes, I think that's a good way to describe them. Again, this is a fairly small minority. Most people, myself included, feel there's very strong evidence for the existence of dark matter or something like it. But of course, we need to keep an open mind.
I think the paper here gives a scenario where the standard picture of dark matter and galaxy formation actually explains the observations quite well. There was dark matter surrounding these galaxies, and they got separated from their dark matter during this collision. It will be interesting to see how that's received by various different camps in the field as this goes forward.
This won't be the last word, but I think it's a very interesting and noteworthy addition to the conversation. It provides a compelling picture of how this system could have formed.