Galaxy cluster at the edge of the Universe

By Phil Plait | May 10, 2010 9:57 am

Astronomers have found the most distant galaxy cluster ever seen: the sexily-named SXDF-XCLJ0218-0510.

First, the picture, then the words:


Yikes! What’s all that then?

Okay, first, this picture is littered with stars and galaxies. The galaxies are so far away they’re hard to distinguish from the stars! The dots that have arrows pointing to them are the galaxies that are most likely part of the cluster. The ones with circles have had their distance measured and are known to be part of the cluster for sure. The contour lines represent the detection of very hot gas, which is a dead giveaway that we’re dealing with a cluster here; all big clusters have gas swirling around them that gives off X-rays; the lines are like a topographic map telling you where the (otherwise invisible) gas is in the picture.

"So what?", you might say. We’ve seen lots of clusters before. Ah, but this one is different: it’s a whopping 9.6 billion light years away.

Billion. With a B.

Studying clusters is important because it tells us a lot about the environment in which galaxies form. When we’re trying to figure out the overall structure of the Universe — and it’s incredible to me that we can even think about doing such a thing — clusters are the single biggest component. Understanding them means understanding the Universe itself.

But there’s more. Because clusters are so big and bright, they can be seen really far away. In space, distance means time; the farther away we see an object, the younger the Universe was when the light left that object. In the case of this newly found cluster, the light we see left it 9.6 billion years ago — making it 400 million light years farther away than the next-most distant cluster ever seen. The Universe itself is only 13.7 billion years old, so we’re seeing this structure as it was not too long after it formed.

And that’s the key issue. Clusters are so big that they take quite a bit of time to form. But how long? We’re not sure. We’re not sure exactly how long galaxies take to be born either. So every time we see one farther away we push back the time it takes to form them. Think of it this way: if it takes, say, 5 billion years to form a cluster, then we wouldn’t see any more than 13.7 – 5 = 8.7 billion light years away. This one is 9.6 billion, so we know that clusters cannot take more than 13.7 – 9.6 = 4.1 billion years to form. In reality they probably take quite a bit less time. Observations like this one will help us understand just how much less.

The reason this is important is because we don’t know the exact timeline of the Universe after the Big Bang. We know when it happened, and we know when the first stars formed, but it’s hard to say when the first galaxies and clusters of galaxies started to come together. Most likely that wasn’t a firm time, but it was spread out over hundreds of millions or billions of years. But the more we see, the better we can figure that out. As it happens, the colors of the galaxies in this image give a clue as to how old they are: young stars are blue, and old stars are red, so by looking carefully at the mix the age of the galaxies can be estimated. The galaxies in this cluster look like they formed around 11.5 billion years ago, making them already a couple of billion years old when they emitted the light we see in the picture.

Another interesting thing: you can’t really tell from this image, but in the original data it looks like there are actually two clusters here at about the same distance. One of them has that hot gas, but the other is more indistinct and the detection of gas isn’t as strong. It’s entirely possible that this other cluster isn’t quite fully formed yet. If that’s so, then that means astronomers are able to detect clusters so far away that we can now actually see them forming. The more of those we find, the better our understanding will be of just how these enormous structures came to be.

Looking at distant clusters also tells us about the early universe. You can’t build a skyscraper unless the ground can support it, and you can’t build a galaxy cluster unless the Universe has the right conditions for it. The Universe was different 10 billion years ago: smaller, denser, warmer. All of these are important parameters when you’re building something like a cluster.

So by examining the edifice we can understand the environment it sits in. In this case, the building is a thousand galaxies each with a hundred billion suns, and the environment is the cosmos itself. It’s an architectural study written across the entire sky! And every detail gets us that much closer to understanding how everything — literally — came to be.

Image credit: Subaru/XMM

CATEGORIZED UNDER: Astronomy, Pretty pictures, Science

Comments (56)

  1. Theron

    The real story of course is that we must be wrong about the speed of light, since the universe is only 6000 years old.


  2. @Theron:

    No no, you have it all wrong.

    The universe is many many billions of years old, it’s only Earth as it is that’s 6000 years old. In the beginning, God made the heavens and the Earth. Then he puttered about for a few billions of years before remembering to do something with Earth. You know how it is, you start a project, get distracted, and come back a little later to finish it up.

    And on the seventh day, God had a beer.

  3. dWhisper

    Part of me wants to post “but this is an astronomy website! This should be the only post!” but I can just glance at the threads even slightly off topic for that.

  4. Theron

    Thanks Kevin@3 – now it all makes sense!

  5. Astronomer

    This galaxy cluster was actually first discovered by a different group two months earlier. Their paper is at:

  6. @Theron (4):

    No problem. Also, on the third day, God had to make a quick run to the World Depot because the trees weren’t the right size for the planet.

  7. brainwise

    I love this post by Phil, but it is Kevin whom I shall be quoting for some time to come (“And on the seventh day, God had a beer”). Well played!

  8. @ Theron: Shouldn’t that be Universe Depot? Of course, he probably also picked up some fertilizer for the Garden of Eden while he was there.

  9. Jim

    And seeds for a tree that he was promised was totally, totally harmless.

  10. andrew

    the universe was created because of an explosion in a meth lab. george carlin

  11. Adrian Lopez

    Does this cluster have cows that want to be eaten and are capable of saying so?

  12. shawmutt

    Brain…hurting…must…try some cohesive…thoughts.

    Alright, dumb question. Isn’t this thinking Earth-centric? If we’re measuring waves of energy hitting the earth, isn’t that telling us the age of the universe relative to the age of Earth, not relative to the beginning of the universe? How do we know we’re looking at the edges?

    I tried to look at the wiki article on the age of the universe but nuerons started smoking. edit: I can’t even spell nuerons anymore…

    Dammit Jim, I’m an aspiring biologist, not an astrophysicist!

  13. squib

    @shawmutt 12.

    it’s impressions of part of the universe as it was 9.6 billion years ago. That light we’re seeing, it ain’t from yesterday.

    if we find something 13 billion light years away, then we’ll get a closer look at what the universe looked like shortly after the big bang (provided you believe in that sort of thing).

  14. Sir Eccles

    Stop writing about Good Astronomy, that’s two in a row. I came here to read about Bad Astronomy and I am very disappointed!

  15. Gustavo Lozano

    Earth is no 6000 years old, actually it is more than 4.5 BILLION YEARS OLD.

  16. Jim2

    Red shift only provides the velocity of an object on an axis toward or away from the observer. Is there a method for determining, estimating, or guessing lateral movement of galaxies or stars? Other than waiting a few million years and looking for positional change? Is the Andromeda galaxy really going to intersect the Milky Way or do we just know it is sort of heading in our direction and might miss the Milky Way by a million light years?

  17. I started to write something up but indeed my brain started to smoke as well. Give us more distant objects, more more more! :)

  18. Bob Sireno

    Old galaxy clusters are awesome. If we were able to see back past the all the matter and past the cosmic microwave background radiation to an invisble (as of today) point 13.7 billion years old wouldn’t every view of it in the sky see the exact same cosmic embryo, but from a different angle?

  19. @shawmutt 12

    You bet it is Earth-centric; we haven’t found a way to view the universe from the POV of Andromeda. Still, there are ways to nail down the age of the entire universe, the best of which is the Cosmic Microwave Background. By knowing how much stuff is in the universe and how the universe is shaped, we can use measurements from the CMB to get an accurate measure of the age of the universe, independent of it’s size. Because we don’t know the size of the universe, all we can really talk about are the areas we can see, which amount to a sphere of radius 13.4 billion light years centred around us.

  20. Theron

    Gustavo – joke, joke, yes. We keeed. Really. :)

  21. Question: Would these galaxies even exist today, or would the stars have all “burned out” their fuel?

  22. Paul

    If the accelerating expansion of the universe is due to a non-zero cosmological constant, is this distant cluster already causally disconnected from us (meaning: no signal we send now could ever reach it, or vice versa)?

  23. Beautiful, but why did they name it after a licence code for MS Office or Adobe Photoshop?

  24. Drewwerd

    All of these comments relate to what we know as the LOCAL universe (somehow the uni- part is what’s a bit self-righteous) nearly all of the Standard Model and related theories are suffcient for some version of what we see in this 13.5 BY ‘bubble’, but they do not disallow other bubbles. Moreover, the other bubble-verses could have different laws of physics. That should get yer neurons cookin’!

  25. Navneeth


    I suggest you take a look at Ned Wright’s Cosmology Tutorial.

  26. blax

    “…if it takes, say, 5 billion years to form a cluster, then we wouldn’t see any more than 13.7 – 5 = 8.7 billion light years away..”

    Isn’t the size of the universe supposed to be way bigger than 13.7 GLY ? Could something created 8.7 GY ago be farther than 8.7 GLY away because of expansion of space (can’t remember the word for this)??

    I know I’m mistaken. But don’t know where….

  27. Jose

    Paul, from

    you can get the cluster’s comoving distance, which is about 15 Gly. From Figure 1 in

    you can check that the cluster is about to become causally disconnected (but has not done so yet). This squares with

    where the unreachable region begins at z=1.69, whereas the cluster’s redshift is 1.62.

  28. Wayne Robinson

    “Alright, dumb question. Isn’t this thinking Earth-centric? If we’re measuring waves of energy hitting the earth, isn’t that telling us the age of the universe relative to the age of Earth, not relative to the beginning of the universe? How do we know we’re looking at the edges?”

    The way I understand it (and it makes my neurones hurt too; I’m only a lowly retired pathologist), every point in the universe is the centre, because 13.7 billion years ago, all space was concentrated in something very close to a singularity, and with the Big Bang, it was space that “exploded”, creating matter with vacuum quantum fluctuations, and therefore dragging matter with it. So every point in the Universe we can see comes from that point and is therefore the “centre.”
    Space is still expanding; the greater the distance, the greater the velocity of separation, so at a great enough distance, the speed of separation is greater than the speed of light. The space between us and Andromeda is expanding at a much lesser speed, but because Andromeda is so close (“only” about 2 million light years), gravity will still cause us to collide with it (I’m looking forward to seeing it in a few billion years).

  29. Torbjörn Larsson, OM

    I wonder what that franchise serve, Solar Steak with Radiation Fries?

    [Kidding, it is truly edgesome.]

    Is there a method for determining, estimating, or guessing lateral movement of galaxies or stars?

    In the spirit of shawmutt’s biology, I could suggest Lateral Mass Transfer. The pun aside, I don’t see why you wouldn’t be able to use such mechanical “genetic” properties such as embodied in the virial theorem (and others) to supplement statistics of doppler measurements with knowledge of the mass distributions and where they take bodies.

    For individual cases, I dunno, but astronomers would know if it can be done.

    wouldn’t every view of it in the sky see the exact same cosmic embryo, but from a different angle?

    If I understand the hyperbolic geometry correctly, what you see as you look further out is an ever dimming and ever denser packing of stars or whatever mass you use for emitting light to make your observations from. Eventually you will see nothing, have too few photons.

    What happened before the putative inflation epoch is uncertain but bears on this. Anyway, the “embryo” as you term it isn’t necessarily confined to the same “volume” of “worldlines” or whatever you have before them, even if it come out of a Planck volume in some theories.

    [Note added in posting: Wayne Robinson’s comment complement but describes this better, no “volume centric” view.]

    In any case, it isn’t a well defined way of describing the physics, I believe. We aren’t observers “outside, looking into a volume”. We are inside the volume, so there is no angle from which to “look into”. The attempted analogy breaks down before the physics does.

  30. Joey Joe Joe

    Isn’t asking “when did clusters form” a bit like asking “when did this puddle form”? I mean, at what point do those few drops of water on the ground become a puddle?

  31. jcm

    How does it stack up (hah!) against the Hubble Ultra Deep Field (HUDF)?

  32. lewis haymes

    why do astronomers et al. use the phrase ‘universe’ when discussing things of this sort. Shouldn’t the phrase be ‘known universe’?

  33. Ford Prefect

    MichaelL, good question. Which in turn asks the next question, how easily, and how fast, can a spiral galaxy turn into an elliptical galaxy? Because, spiral galaxies have a lot of spare dust and gas to make stars with, while an elliptical one doesn’t, and consequently has done with star formation.

    Joey Joe Joe, I think the answer to that is “how deep is the depression the water is in?” Or in cosmological terms, “what are the characteristics of the cloud of gas the galaxies are embedded in?” If the depression’s shallow, I don’t think you’ll get a puddle; if the gas is far too thinly spread, you won’t get a galaxy cluster – as the most you’ll get a heap of spread-out low-intensity galaxies.

  34. HyLite

    So… The experiment was created, the question asked and they looked. The wave collapsed and behold it was created with all its 9.6 billion years of history.

  35. Of course the use of the word universe all inclusive is speculative and theoretical, but the common use of the word of course refers to the observable universe. One I think can read between the lines concerning the meaning of the word universe in a particular context.

    I believe the question that should be asked is : how did such a large and complex galaxy cluster evolve in such a relatively short period of time as presumed by present theory. The answer to this question I think is very simple: it cannot. In simple language it could be that the universe is far older than present theory allows.

    As time progresses and observational capabilities increase I believe this possibility will become crystal clear. The sooner this is realized the better, in my opinion.

  36. DaveH

    I believe the question that should be asked is : how did such a large and complex galaxy cluster evolve in such a relatively short period of time as presumed by present theory. The answer to this question I think is very simple: it cannot.

    Why not?

  37. >Why not?

    because it take time for matter to assemble into clusters – have a look at the growth of structure in some n-body cosmological models – clusters arrive late on the scene

  38. Gary Ansorge

    19. HJ Hornbeck

    ” By knowing how much stuff is in the universe ”

    ,,,and until very recently, we only knew about 4 % of the total mass energy of this universe, so I wonder how/if that other 96% has been adequately incorporated into those calculations?

    35. HyLite


    Gary 7

  39. Pi-needles

    Billion. With a B.

    Hey, we know how to spell ‘billion’ – its not like we’re fat-fingered Wall street bankers here! 😉

  40. JB of Brisbane

    T. Larsson says: “I wonder what that franchise serve, Solar Steak with Radiation Fries?”

    Actually, it’s Nuclear Fission Chips.

  41. Tim

    If we are looking at something that is 9.6b light years away, could someone on the other end do the same to us (with equal technology or better) and assume we, too, are 9.6b light years away, making US that old, or older? Or would we just not show up?

  42. Paraducks

    Question for Theron. If God created the heavens and the earth, then god would obviously be in a position to look down on his creations and monitor them; that being said why would god use our Sun to measure his days? One of god’s days (as paraphrased by the most coporeal God that I know of, George Burns) must be much longer than one of earth days for that to work out. And at which point does that intertwine with man days, biblically speaking?

  43. Jon Hanford

    How quickly we forget. Last year, the galaxy cluster JKCS 041 was reported to have a redshift z=1.9 (10.2 billion light-years), thus making it the most distant galaxy cluster known[ ]. It should be noted that this is a photometrically-derived(i.e. less accurate) redshift and not a spectroscopically-derived value (as is the case in this z=1.62 galaxy cluster). Hopefully spectra of this cluster will be obtained soon to help nail down (or refute) its cosmological distance. Sound like a case for Hubble.

  44. Paraducks

    I often wonder about distance measurement in space. On paper it seems straightforward. But in reality I have seen interstellar images of duplicate galaxies and stars in the same image. I cannot remeber the theory that it is based on but it plays into the theory that the universe is actually round and the way that space and time refracts around the edge or gravities of the universe/galaxies/black holes causes a doppleganger effect in optics (gravitational lensing?) showing multiple images of the same galaxy or star to apprear in the same image. I know they use heat signatures and radio waves among other things to gather information about galaxies and stars but what I’d like to know is what the margin of error is on this kind of data since it is mostly based on theories and what we think are solid mathmatical laws?

  45. @Paraducks (43)

    You left this yesterday – but, umm.

    It was a joke, I’m sure.

  46. Brian

    Okay, so I am not a physicist, but something is not quite linking up with me. Lets say the known universe is 13.7 billion years old. If this cluster is 9.6 billion light years away, then what we are really saying is that the light we are seeing is 9.6 billion years old. Got it.

    Here’s the problem that confuses me. 13.7 – 9.6 = 4.1 billion years. Somehow, during that 4.1 billion years the Earth and this cluster managed to separate by 9.6 billion light years? If the Earth and this cluster are both made of mass, then neither can travel faster than the speed of light. I figure that if we are heading in completely opposite directions, we must travel less than 2x the speed of light with respect to eachother, right? So let me put the broken thoughts of my head together: IF the cluster existed at the exact moment of the Big Bang and IF we are traveling in exactly the opposite direction from the center of the universe and IF we are both traveling at 100% the speed of light, this cluster could still be no more than 8.2 billion light years away? That’s a lot less than 9.6. To make things worse, if we were moving apart at a full 2x the speed of light, then the light that left the cluster would NEVER reach us

    Help me out here. I can’t be the first person to consider this. Surely there is someone here who can explain? With that, I think I am going to join God and grab a beer. Who knows, if a day is so much longer for Him, maybe He is still drinking it and just relaxing by the “primordial” pool.

  47. Brian

    Nevermind. A six-pack and a good, simple web page soothed my worries.

  48. The bad astronomy is how distances are usually reported in the popular press. The real observed distances are far, far greater than 9 or 10 billion light-years. For example, if you look at the Supernova Cosmology Projects data on Type 1a SN, the furthest such star is at a distance of about 38 billion light-years (converting from SCP’s distance modulus units which are based on brightness and the inverse square law). And the furthest galaxies are at much greater distances than that. The usual explanation for how the universe could be so large is of course that expansion has occurred at speeds much greater than c.

    “Distances obtained as the speed of light multiplied by a cosmological time interval have no direct physical significance.” –Wikipedia article:

  49. John Tomassoni

    It seems to me that since it takes time for light to reach us, then the farther we see in space, the farther back in time we see — we are getting closer to the beginning of things. This appears to be the reason why we cannot or never will see the ‘edge’. So the object observed had to be closer to the beginning, not close to the edge.

  50. Old wise man

    I see I’m a little late but… Remember Einstien said that the cosmological constant was his biggest blunder and now people are saying that it may have been his greatest achievement? Well maybe they are both right (sort of.) He definitely did not like the big bang idea. In my thoughts this could actually be explained with elements of the big bang theory and steady state theory.
    Multiple big bangs would be what some would call it but that doesn’t give the correct idea. Perhaps spacetime is infinite in age and we are seeing different “big bangs” because the “universe” is infinite and steady creation is also occurring. It takes some thinking but it would take steady state theory of constant creation and evolution and marrying it with ever faster expansion believed by big bang theorists. Effectively an infinite age of the universe and also increasing expansion. Its increasing expansion rate just means that we see less of the older stuff at the edge of what light can reach us because it is surrounded by more new stuff.
    The observer is at the center of its creation, but creations start all the time all over. It seems that both of the 2 theories have to come up with more and more arguments as to why that theory is correct, and the explanations seem to agree with each other more and more! Neither is correct and a new theory is on the way which will probably not be correct unless we can find the answer to combining quantum and relativistic theory, if then.

    I think Einstien had the idea that God is infinite, which would join with this type of explanation, but he nor anyone so far can actually explain how it works and likely we never will.

    Perhaps someone involved with fractal theories could explain it better and come up with the answer! 42!

  51. Old wise man

    One extra thought. Time is relative and this must be factored into the explanation I gave if you didn’t blow your neurons already!

  52. “This one is 9.6 billion, so we know that clusters cannot take more than 13.7 – 9.6 = 4.1 billion years to form. In reality they probably take quite a bit less time. Observations like this one will help us understand just how much less.”

    This reasoning is only valid as long as the Big Bang (BB) is a valid theory. If you take away the BB limit on time (about 14 billion years) I think you will find the truth. Not only is this a galaxy cluster ~9.6 billion light years away, but is an old looking cluster accordingly with little perceptible stellar production. I think this is a very strong indication that the universe is much older than the present BB theory could allow. This is not an isolated observation but it is the best one concerning a large galaxy cluster at this distance/ redshift. There have been many such observations and claims of old appearing galaxies and clusters at great distances since the Hubble has been making such observations. I have a collection of such papers. I think many more such observations will be made after the James Webb goes up.

    The Milky Way is thought to be 12 billion years old based upon the age dating of some of its oldest stars. On an average It does not appear to be an old looking spiral galaxy based upon its average constituent stars; instead it may be a typical example of a middle age spiral galaxy. The subject cluster is believed to be made up of primarily old appearing galaxies. I believe all of such observations collectively have for maybe 10 years now, been harbingers and strong evidence that the Big Bang idea/ theory may be wrong.

  53. Bryson

    Well said Forrest, I think that we do science a huge dis-service when we use anything (not just theory, but law as well) as an absolute. We should always say that we think that the universe is 13.7 billion years old, never that IT IS 13.7 billion years old. It is fine to use previous data, theory, conclusions and laws as a foundation to further understand the next unknowable thing we soon make knowable, however we must never make these previous accomplishments a concrete non-flexible wall that we put up that doesn’t allow us to see the new information clearly. When we thought the earth was flat we would not listen to reason that the earth might not be what we thought it was. If we take the BB theory and make it unchangeable, unbend-able, undeniable fact we may never see the universe as round, we might just keep on seeing it as flat. (I say we see it as flat now because unless we got REALLY lucky, we guessed wrong as to what the universes history is.)


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!


See More

Collapse bottom bar