Skip to content
Starts With A Bang

Ask Ethan: Why will the Milky Way and Andromeda collide?

The Universe isn’t just expanding, the expansion is also accelerating. If that’s true, how will the Milky Way and Andromeda eventually merge?
antennae galaxies NGC 4038 4039
This Hubble Space Telescope image of the Antennae Galaxies, NGC 4038 and NGC 4039, shows two comparably-sized spiral galaxies in the process of interacting and merging. In about 4 billion years, the Milky Way and Andromeda will undergo a similar interaction.
Credit: NASA, ESA, and the Hubble Heritage Team (STScI/AURA)-ESA/Hubble Collaboration
Key Takeaways
  • Although they’re currently separated by 2.5 million light-years, the Milky Way and Andromeda are heading toward each other, and will eventually merge 4-to-7 billion years from now.
  • Overall, not only is the entire Universe expanding, with galaxies spreading out and getting farther apart over time, but the expansion is accelerating, meaning galaxies are speeding up in their recession from each other.
  • How can we reconcile these two simultaneously true facts? If the Universe is not just expanding, but accelerating, then how are galaxy mergers still occurring? Let’s unpack the answer.

Of all the galaxies in the Universe that lie beyond the Milky Way, none looms larger than our “big sister” in the Local Group: Andromeda. Andromeda has more stars, more mass, and a larger physical extent than the Milky Way in all three dimensions. It spans a larger angular extent in our sky than six full Moons all lined up next to one another, and despite its location some 2.5 million light-years away from us, it’s actually moving in our direction, setting up a collision that should happen 4 billion years in our cosmic future. Another 3 billion years later, the greatest galactic merger in our Local Group’s history will be complete, leaving just one behemoth of a galaxy at its core: Milkdromeda.

But why is this happening? After all, not only is the Universe expanding, but the expansion of the Universe is accelerating, too! How could these two seemingly paradoxical points both be true: the expanding Universe is accelerating, but Andromeda is heading toward us and is destined for a collision-and-merger with us? That’s what Robert Asselta wants to know, writing in to inquire:

“If the universe is expanding and galaxies are spreading apart from each other, then why/how is Andromeda expected to collide with the Milky Way a few billion years from now?”

It’s a very smart question to ask, and the answer isn’t necessarily obvious. But if we go through the details, a clear answer emerges. Let’s find out!

Mayall's object
This Hubble Space Telescope image of Mayall’s object, also known as Arp 148, shows two galaxies in the process of collision. As one galaxy punches through the center of the other, stars form in both galaxies, but the one that got “punched” is having its gas propagate outward in waves, triggering new star formation on its way toward creating an overall ring-like shape. While they interact and merge, galaxies can take on many fascinating and peculiar shapes.
Credit: NASA, ESA, the Hubble Heritage (STScI/AURA)-ESA/Hubble Collaboration, and A. Evans (University of Virginia, Charlottesville/NRAO/Stony Brook University)

The cosmic race

Ever since the dawn of the hot Big Bang, the Universe has been doing two things relentlessly. On the one hand, the Universe has been expanding, with the rate of expansion determined — at any particular moment in time — by the overall energy density of space, on average. Energy density includes energy in the form of:

  • normal matter,
  • dark matter,
  • radiation (like photons),
  • neutrinos,
  • dark energy,

as well as anything else that might possibly exist, from exotic species of energy to topological defects to spatial curvature to anything present in extra dimensions. If you can calculate the total energy density due to all sources that contribute, plus the effects of spatial curvature, plus any effects due to a cosmological constant, you’ll know the expansion rate of the Universe at any moment in time.

But on the other hand, the Universe, even as it expands, also gravitates, with all forms of energy not just curving the local vicinity of the space they occupy, but affecting the overall expansion rate of the Universe. Recognizing this relationship between the different forms of energy present and the overall behavior of the Universe was first achieved in 1922, with the work of Alexander Friedmann in the context of Einstein’s General Relativity. Although this work is more than a century old, Friedmann discovered all three of the major possibilities that would be expected to occur.

Raisin bread expanding Universe
The ‘raisin bread’ model of the expanding Universe, where relative distances increase as the space (dough) expands. The farther away any two raisin are from one another, the greater the observed redshift will be by time the light is received. The redshift-distance relation predicted by the expanding Universe is borne out in observations, and has been consistent with what’s been known all the way back since the 1920s.
Credit: Ben Gibson/Big Think; Adobe Stock

Our Universe’s overall fate

On the largest of cosmic scales, the Universe behaves as though it’s a race between these two phenomena:

  1. the initial expansion rate that it began with at the start of the hot Big Bang,
  2. and the gravitational effects of all the various forms of energy that exist within that Universe.

The Universe is a race between these two effects, and the onset of the hot Big Bang is the “starting gun” between the only two competitors in this cosmic race. As Friedmann understood it, there would be three possible outcomes.

  • The initial expansion could be too great for the amount of “stuff,” like matter and radiation, that are present in the Universe. In this case, the expansion would win, and although the effects of gravitation would slow the expansion down, the expansion rate would always remain positive, and the Universe would dilute, becoming emptier and emptier, with no end.
  • Alternatively, there could be too much gravitating “stuff” in the Universe for the expansion rate to keep up. Gravitation wouldn’t just slow the expansion rate, but after enough time, it would grind the expansion to a halt. But with all that energetic “stuff” still in it, gravitation would continue, and the Universe would now contract. This reversal of expansion, contraction, would eventually lead to a Big Crunch.
  • Or, just like Goldilocks and the three bowls of porridge, the three chairs, and the three beds, it’s possible that the Universe is “juuuuuust right,” and the expansion rate and gravitation perfectly balance. The Universe expands, but gravitation slows it: so that it approaches, but never quite reaches, zero. If one more atom were present, it might recollapse, but instead, it forever remains expanding, just by the tiniest amount.
dark energy fate
The expected fates of the Universe (top three illustrations) all correspond to a Universe where the matter and energy combined fight against the initial expansion rate. In our observed Universe, a cosmic acceleration is caused by some type of dark energy, which is hitherto unexplained. All of these Universes are governed by the Friedmann equations, which relate the expansion of the Universe to the various types of matter and energy present within it.
Credit: E. Siegel/Beyond the Galaxy

Perhaps the only fault you can find with Friedmann’s analysis, even looking back more than 100 years later, is that he failed to anticipate that one of the categories of “stuff” that would be in the Universe would be a form of dark energy. As it turned out, the Universe really did look like it was on that Goldilocks path for about the first ~7 billion years of cosmic history: with the expansion rate dropping and dropping as gravitation slowed it down. It looked, if you had lived back then and been well-versed in the intricacies of modern physical cosmology, exactly like that “just right” case we outlined above.

But when the matter (both normal and dark) and radiation (and neutrinos) diluted past a certain point, a new effect began to appear: what we call dark energy today. This form of energy behaves as though it’s inherent to the fabric of space itself, so as the Universe continues to expand, it doesn’t dilute the way matter or radiation dilute; even as the volume of the Universe expands, its energy density remains constant.

This changes the fate of the Universe from the third option that Friedmann anticipated — the Goldilocks case — into an extreme version of the first option (the “expands forever” case): where not only does the Universe become emptier and emptier as time goes on, but distant galaxies, as they recede from one another, appear to recede at increasingly faster and faster speeds.

evolution of matter radiation dark energy
How matter (top), radiation (middle), and dark energy (bottom) all evolve with time in an expanding Universe. As the Universe expands, the matter density dilutes, but the radiation also becomes cooler as its wavelengths get stretched to longer, less energetic states. Dark energy’s density, on the other hand, will truly remain constant if it behaves as is currently thought: as a form of energy intrinsic to space itself. These three components, together, dictate how the Universe expands at all times from the Big Bang until the present day and beyond.
Credit: E. Siegel/Beyond The Galaxy

The global fate of the Universe

If you start here in the Milky Way and look at a distant, receding galaxy from us, you’ll find that its light is redshifted: that its wavelength has been stretched by the expanding Universe. As time continues to march forward, you could keep on monitoring the light from that galaxy, and see how it was changing. Would the amount its light was stretched by:

  • increase,
  • decrease,
  • or remain the same,

as it continues to recede but also as the rate of expansion continues to evolve?

If you were watching that galaxy for the first 7.8 billion years of our cosmic history, you would’ve seen that “stretching amount” decrease, corresponding to that galaxy slowing down in its recession from our perspective. If you watched that galaxy when the Universe was precisely 7.8 billion years old, you would’ve seen that “stretching amount” remain the same, corresponding to that galaxy “coasting” in its recession, or continuing to recede at the same speed. And if you watched that galaxy over the most recent 6 billion years of cosmic history, you’d see the “stretching amount” of its light increase with time, implying that it was receding faster-and-faster.

That’s what we mean when we state “the expansion of the Universe is accelerating,” that for the past 6 billion years, any distant object we’ve looked at appears to recede ever faster and faster as time goes on. This is still the case today.

local supercluster
This illustrated map of our local supercluster, the Virgo supercluster, spans more than 100 million light-years and contains our Local Group, which has the Milky Way, Andromeda, Triangulum, and about ~60 smaller galaxies. The overdense regions gravitationally attract us, while the regions of below-average density effectively repel us relative to the average cosmic attraction. However, the individual groups-and-clusters are not gravitationally bound together and are receding from one another as dark energy dominates the cosmic expansion.
Credit: Andrew Z. Colvin/Wikimedia Commons

But what about smaller cosmic scales?

The story we’ve just told about cosmic expansion is strictly true, but it technically only applies to the overall Universe as a whole. The reason that’s the case is because there’s an assumption that went into the Einstein Field Equations — the equations that govern General Relativity — that allow us to make the simplifying assumption that Friedmann himself made back in 1922: that all forms of matter and energy are equally and uniformly distributed throughout the Universe. This is valid on the largest of cosmic scales and, for a typical region of the Universe, it’s valid on average.

But the Universe isn’t actually uniform everywhere.

Instead, the Universe is filled with structure: galaxies, groups of galaxies, rich clusters of galaxies, and vast cosmic voids separating them. When we map it out in sufficient detail, we find a web-like network of structure in our Universe, where galaxies form along strands of that network and — more richly — at the nexus, or intersection, of those various strands. Matter gets preferentially attracted into these overdense regions, which causes it to flee the “in-between” regions, creating vast cosmic voids, with the difference between structure-rich and structure-poor regions growing more severe as time goes on.

dark matter-free
The cosmic web that we see, the largest-scale structure in the entire Universe, is dominated by dark matter. Simulations of the large-scale structure of the Universe must include both dark matter and normal matter, including the effects of star-formation, feedback, and gas infall, as all of them are needed in order to predict the emergence of visible structures. Identifying which regions are dense and massive enough to correspond to star clusters, galaxies, galaxy clusters, and filaments, as well as determining when and under which conditions they form, is one of the great achievements of modern cosmology.
Credit: Ralf Kaehler/SLAC National Accelerator Laboratory

The reason for this goes all the way back to the hot Big Bang itself. It turns out that sure, on average, the Universe is filled with the same amount of all forms of energy, including normal matter and dark matter, everywhere. But the truth is that the Universe was born seeded with tiny imperfections: overdense and underdense regions, at the level of just a few parts per 100,000 everywhere.

  • Where you have an overdense region, the more successful you are at attracting more and more matter into you, and the more likely you are to grow to become some sort of massive structure: a star cluster, galaxy, group of galaxies, or even a rich cluster of galaxies, depending on the magnitude and physical extent/size of your overdensity.
  • Where you have an underdense region, the more likely you are to give up your matter to a nearby dense region, and to expand and dilute into a diffuse cosmic void.

In reality, the Universe is filled with both types of regions on all cosmic scales, and those regions grow and shrink according to the laws of gravity, the expansion of the Universe, and whatever happens to be going on around them.

dark matter
This snippet from a structure-formation simulation, with the expansion of the Universe scaled out, represents billions of years of gravitational growth in a dark matter-rich Universe. Over time, overdense clumps of matter grow richer and more massive, growing into galaxies, groups, and clusters of galaxies, while the less dense regions than average preferentially give up their matter to the denser surrounding areas. The “void” regions between the bound structures continue to expand, but the structures themselves do not.
Credit: Ralf Kaehler and Tom Abel (KIPAC)/Oliver Hahn

Who wins?

The only reason we have any structure at all in the Universe — things like star clusters, galaxies, galaxy groups, and galaxy clusters — is because there are regions, locally, where enough matter has accumulated together for gravitation to win: to “win” so thoroughly that it can indeed overcome the expansion of the Universe.

Travel the Universe with astrophysicist Ethan Siegel. Subscribers will get the newsletter every Saturday. All aboard!

The way this occurs has been studied in detail by the branch of science known as physical cosmology, which deals, in part, with the formation of large-scale structure in the Universe. In the early stages of the Universe, overdense regions only grow slowly relative to the cosmic average.

  • By the time 1 million years have passed since the Big Bang, the densest regions are only about ~0.1% denser than the average density.
  • By the time 10 million years have passed since the Big Bang, the densest regions may only have grown to be about ~10% denser than the average density.
  • But after a few tens of millions of years, the very densest regions have now reached a critical point: where they’re about ~68% denser than the average density.

Once that point is reached, something very important happens: gravitation is now important enough that the expansion of the Universe starts losing to gravity in this region of the Universe. Gravitational collapse becomes all but inevitable, and a bound structure will form.

how much dark matter
While the web of dark matter (purple, left) might seem to determine cosmic structure formation on its own, the feedback from normal matter (red, at right) can severely impact the formation of structure on galactic and smaller scales. Both dark matter and normal matter, in the right ratios, are required to explain the Universe as we observe it. Structure formation is hierarchical within the Universe, with small star clusters forming first, early protogalaxies and galaxies forming next, followed by galaxy groups and clusters, and lastly by the large-scale cosmic web.
Credit: Illustris Collaboraiton/Illustris Simulation

This occurs on small cosmic scales first, leading to star clusters: likely when the Universe is only between 100-200 million years old. Then it occurs on larger scales: with star clusters merging together and larger cosmic scales collapsing to form galaxies: likely when the Universe is a few hundred million years old. Then, still larger scales collapse, leading to the first galactic groups and the earliest proto-clusters of galaxies: within the first ~1 billion years of our cosmic history. And finally, you only get mature galaxy clusters after a few billion years have passed, owing to the enormous cosmic scales (and the limit set by the speed of light) at play.

The reason Andromeda and the Milky Way will someday merge together — and yes, they really are on a collision course — is because way back in the early stages of the Universe, more than 10 billion years ago, we all got gravitationally drawn in to become part of the same gravitationally bound structure: our Local Group. Eventually, given enough time, all of the galaxies within our Local Group will collide and merge together, although this process should take several tens of billions of years, multiple times the present age of the Universe, to complete. The Milky Way and Andromeda should approach each other for the next 4 billion years, begin merging at that time, and complete their merger after about another 3 billion years: a total of 7 billion years from now.

milky way andromeda merger
A series of stills showing a visualization of the Milky Way-Andromeda merger and how the sky will appear different from Earth as it happens. This merger will begin occurring roughly 4 billion years in the future, with a huge burst of star formation leading to a depleted, gas-poor, more evolved galaxy ~7 billion years from now. Despite the enormous scales and numbers of stars involved, only approximately 1-in-100 billion stars will collide or merge during this event. The final form of the galaxy, despite the illustration here, is more likely to be a gas-rich, disk-possessing galaxy than the elliptical one shown, as only a small percentage of major mergers lead to a red-and-dead, gas-free elliptical final state.
Credit: NASA; ESA; Z. Levay and R. van der Marel, STScI; T. Hallas, and A. Mellinger

The only reason that merger will occur, however, is because the Milky Way and Andromeda are already part of the same gravitationally bound structure, the Local Group, that became dense enough early enough on to overcome the expansion of the Universe. Although the Universe began accelerating 6 billion years ago, there were still enough dense, growing regions that were gravitationally attracting one another and the matter around them that structures had until about 4.5 billion years ago — around the same time that the Sun, Earth, and Solar System were forming — to become gravitationally bound.

Today, the situation has already been decided: if you’re part of a gravitationally bound structure, you’ll wind up bound to it; if you haven’t yet gotten there, you never will be. Although the Milky Way, Andromeda, and all of the remaining Local Group galaxies will eventually merge together, our Local Group itself will never merge with any of the galaxies, galaxy groups, or galaxy clusters found outside of it. Our Universe has become like a sea of islands, where each island remains as a solid mass, but where separate islands eternally receding from one another in the vast, accelerating, expanding cosmic ocean.

The only reason the Milky Way and Andromeda will merge is because they became gravitationally bound to one another before dark energy took over. Galaxies will still merge for tens of billions of years to come, but only in those groups and clusters that became gravitationally bound billions of years ago.

Send in your Ask Ethan questions to startswithabang at gmail dot com!


Up Next