JWST takes a peek at the first ever galaxies

Title: Panic! At the Disks: First Rest-frame Optical Observations of Galaxy Structure at z>3 with JWST in the SMACS 0723 Field

Authors: Leonardo Ferreira, Nathan Adams, Christopher J. Conselice, Elizaveta Sazonova, Duncan Austin, Joseph Caruana, Fabricio Ferrari, Aprajita Verma, James Trussler, Tom Broadhurst, Jose Diego, Brenda L. Frye, Massimo Pascale, Stephen M. Wilkins, Rogier A. Windhorst, Adi Zitrin

First Author’s Institution: University of Nottingham, Nottingham, UK

Status: Accepted to The Astrophysical Journal Letters, available on arXiv

Ever since the first data release of the James Webb Space Telescope (JWST) in July, it has become clear that this telescope is going to completely transform our view of the distant Universe. Galaxies that looked like featureless blobs when viewed through the Hubble Space Telescope can now be resolved in incredible detail (see Figure 1), despite the fact that Hubble has been one of the world’s leading telescopes for the past 30 years.

Figure showing four pairs of galaxy images. The top image of each galaxy is faint and blurry, with few features visible. The bottom image of each galaxy shows a galactic disk and spiral features, which were not previously visible.

Figure 1: Four galaxies from the SMACS 0723 field (the focus of today’s paper), as seen by the Hubble Space Telescope (top row) and the James Webb Space Telescope (bottom). Each one displays features that were undetected with Hubble, but can easily be seen with JWST. Credit: NASA/ESA/STScI.

Being able to measure the shapes of galaxies (known as their morphology) is vital if we want to understand how galaxies, including our own, were formed. Galaxies typically come in two shapes: thin, delicate disk-shaped galaxies, and spheroid-shaped elliptical galaxies, but it is still not really clear how and when these different galactic structures emerged. Today’s paper uses early JWST observations of a large galaxy cluster, called SMACS 0723, to measure the shapes of very distant galaxies. With this exciting new data, the authors hope to expand our knowledge of galaxy evolution all the way to the very dawn of our Universe.

Zooming in on the first galaxies

This photo of SMACS 0723 is one of the first images to be released from JWST. The cluster is located about four billion light years away at a redshift of 0.4, but today’s paper actually looks at even more distant galaxies, in the background of this image — many of these have been magnified by the gravitational lensing of the cluster. Specifically, it looks at 280 background galaxies at redshifts between 1.5 and 8, meaning we are seeing them just 1-4 billion years after the beginning of the Universe.

The authors firstly measure galaxy shapes using quantitative properties of galaxies, such as their concentration and asymmetry. Their really exciting findings, however, come from classifying these galaxies by eye, splitting them into three categories: disks, spheroids, and “peculiars”.

Galaxies in this third class have an irregular shape, which can be caused by processes such as starbursts or tidal interactions. Alternatively, collisions between galaxies (known as “galaxy mergers“) that are currently in-progress can lead to these “peculiar” galaxies. These violent events are thought to play a major role in galaxy evolution: in the early Universe mergers allow large amounts of mass to clump together, which can later form a galactic disk. Later on, they can destroy these fragile disk structures, turning disk galaxies into featureless ellipticals.

It turns out that at high redshifts (between 3 and 6), about half of galaxies have a disk shape (Figure 2). This is much higher than we previously thought — the data from the Hubble telescope shows that it found a disk fraction of less than 10% at similar redshifts! Interestingly, according to JWST, the disk fraction also stays roughly constant across the whole range of redshifts.

"Three scatter plots, showing fraction of spheroids/disks/peculiars (increasing from bottom to top) against redshift (increasing from left to right). Data is shown for HST and JWST. HST shows a consistently high spheroid fraction, and that disks decrease with increasing redshift, while peculiars increase. In contrast, JWST shows an increasing spheroid fraction, a consistently high disk fraction, and a peculiar fraction that decreases at high redshift.

Figure 2: Fraction of spheroid, disk, and peculiar galaxies at different redshifts, measured with JWST in today’s paper, and with Hubble (HST) in previous work. The trends found by Hubble had predicted the number of disks would decrease at redshifts greater than three, and that most galaxies would be peculiar. JWST shows that this is not the case. Figure 4 in today’s paper.

A less turbulent Universe?

Our current idea that mergers assemble galaxies in the early Universe means that we would expect to find lots of peculiar galaxies and few disks at high redshift, as these disks are still in the process of forming. However, the near-constant disk fraction found in this study indicates that disk galaxies (like the Milky Way) have existed in a fairly stable state for more than 10 billion years, seemingly contradicting our old ideas.

So what’s going on? There are several ways to interpret these results. It could be that almost all mergers occur extremely early in the Universe, quickly forming disk galaxies, and that these disks survive until the present day because recent mergers are far less common than our current theories suggest. Alternatively, it could be that only some classes of galaxies are built up by mergers, or even that mergers are simply far less likely to destroy disk structures than we previously thought.

Whatever the case, it indicates that we may need to refine current theoretical ideas about how galaxies assemble and evolve through mergers, which is one of the key predictions of our widely-accepted model of the Universe (the Lambda cold dark matter, or ΛCDM, model). Some articles based on this work have gone a step further, stating that this research disproves ΛCDM, or even the Big Bang. However, despite the homage to noughties emo-pop in the title of this paper, there’s really no reason to panic. Tuning and re-tuning theories to fit new data is a normal part of the scientific process. In fact, this paper is exciting: it tells us that we still do not truly know where galactic structure came from, but that new science carried out using this new telescope will finally give us a chance to understand the origins and lives of galaxies.

Astrobite edited by Aldo Panfichi

Featured image credit: NASA/ESA/STScI

About Roan Haggar

I'm a PhD student at the University of Nottingham, working with hydrodynamical simulations of galaxy clusters to study the evolution of infalling galaxies. I also co-manage a portable planetarium that we take round to schools in the local area. My more terrestrial hobbies include rock climbing and going to music venues that I've not been to before.

Discover more from astrobites

Subscribe to get the latest posts to your email.


  1. It seems very odd that in all the theories there are none that consider the Universe as living and growing. Such an answer seems natural and logical, yet it appears never to be considered?

    • I’m not sure what you mean by “living”, but it’s certainly true that the Universe is growing. We are only able to determine the distances to these galaxies by measuring their redshift (i.e. the speed at which they are moving away from us), and the reason they are moving away is because the Universe is expanding and growing. We’re actually quite fortunate to live in an expanding Universe — if we lived in a static (not-growing) Universe, we would struggle to measure how far away are these distant galaxies. Einstein was actually once a firm believer in a static Universe, although over his career he changed his mind as more evidence emerged for an expanding Universe. Hope this helps!

      • Roan, thanks for “Astro-splaining” all this, as no matter how many times I read these type of articles I just can’t wrap my head around the space-time expanding universe concept, so please bear with my
        So where did the universe begin I wonder? Vector-distance-time could equal a location, eh? Even with everything coalescing to a single point “at the beginning” that still had to happen somewhere I would guess.
        Has anyone pin-pointed the location of where all this “stuff” occurred/began? Seems there could be smoking gun evidence at this spot?
        Your thoughts??

        • Hi John — I think I’ll start using the phrase “astro-splaining” myself, I like it! I’m not an expert on Big Bang physics, but I’ll try my best to explain.

          The Big Bang is difficult to get your head around because it didn’t really happen in a particular place, instead it kind of happened everywhere at once. Before the Big Bang there was no “space”, and the Universe had no volume. Space itself was formed when the Big Bang occurred, so the entire Universe was experiencing the Big Bang at the same time.

          Because of that, there isn’t a spot in the present-day Universe where the Big Bang happened, because it happened everywhere! You can perhaps visualise this if you imagine rewinding our Universe, which is currently expanding. If you rewind right back to the beginning, everything will get closer and closer together until it is all in the same place. This means that every place in our present-day Universe was the location of the Big Bang.

          I hope that makes sense — I know a lot of professional astronomers who struggle to get their heads round this, it’s a strange and abstract thing to try and imagine! Hope this helps though.

          • i believe before the big bang there was a supermassive black hole where all our universes materials were pulled into. then after too much material it exploded like a star and dumped our universes building blocks into the vacuum of space. before that black hole was another universe that went through the same process. explosion, formation, expansion, then retraction brought back by gravity or reversal in black matter…wash rinse repeat.

          • Hi John — I think what you’re describing is related to an idea called the “Big Bounce”, where our Universe was preceded by another Universe, that eventually collapsed under gravity and was then followed by the Big Bang. It’s a difficult idea to prove or disprove, but definitely a fun one to think about!

          • Hey Roan excellent explanation to your thoughts and theories as the universe expands after the big bang nearly 14 billion years ago I wonder as of now the universe expanded I’ll say in the hundred billions now and there’s billions to trillions of objects out there that’s expanding with it. But at the end of the day none of us will never know how far the universe expand not even that but alien signatures of life to close the case on the answers we been looking for since Isaac Newton & Albert Einstein times . Alot of the cosmic nature that’s in our solar system and beyond we will never get the actual 100 percent of the answers we been searching for the universe is very complicated strange place to understand in science fiction

          • You’re right, we may never understand absolutely everything about our Universe. That doesn’t mean that we should stop trying though! In the last 100 years our understanding of the Universe has made unimaginable leaps forward, so who knows how much better it will be in another 100 years!

          • De ruimte was er in aanname niet voordat de oerknal plaatsvond. Er was niets. Ruimte wordt pas gekenmerkt wanneer er sprake is van twee of meerdere lichamen en wat zich daartussen bevindt.

      • Awesome is what that is

      • Hello and thanks for the great read. I’m very curious about this statement “if we lived in a static (not-growing) Universe, we would struggle to measure how far away are these distant galaxies.” You would think that it would be easier if they were static. Again thanks for the great read keep them coming.

        • Hi Kenny, this is a great question! The problem with galaxies (particularly very distant ones) is that we don’t know how big they are, which makes it difficult to measure their distance. Imagine looking at a tree in a field — if you don’t know much about the tree, it’s not easy to tell if it’s a small tree that’s nearby, or a large tree that is very far away. If galaxies were not moving, we would have this same problem with them.

          However, because our Universe is expanding, most galaxies are moving away from us. Additionally, galaxies that are further away from us are receding more quickly. We can measure how quickly galaxies are moving away based on how “redshifted” their light becomes (things moving away quickly look redder in colour). Then, because more distant galaxies are moving away more quickly, this speed can be used to infer how far away a galaxy is! I hope this helps.

      • Yo Roan, what I mean by living is that matter/energy was not all created in a big bang. While we can see that space is expanding, this proposition is vastly more difficult to quantify, but that does not mean its invalid.
        But such a theory has no trouble accommodating the fact that types of galaxy are consistent across red shifts. It simply means that there was not one big bang, but an ongoing big bang type of process. One measurement that would be handy is energy/mass density across space as it expands. Does it also remain unexpectedly consistent, or at least does not reduce as expected?

        • Dark energy is similar to what I think you’re describing. The density of matter in the Universe descreases over time, because the volume of the Universe increases but the amount of matter stays (roughly) the same. However, the density of “dark energy” stays the same, even as the Universe expands. We still know very little about dark energy and how it works, but we know that it acts against gravity, pushing the Universe apart. It is therefore causing the expansion of the Universe to accelerate, even billions of years after the Big Bang!

    • First ever Galaxies LMAO really . So when we build a better Telescope and see past those Galaxies are you Scientists going to call those the First Ever . Surprise there was no Big Bang it’s just a Satire written in a News paper along time ago .
      The farther we look we will continue to see More there is no end . We are just Cosmic dust on field with no being no ending.

      • Hi Nathan, thanks for commenting — there’s a huge amount of evidence for the Big Bang, from the Cosmic Microwave Background (the radiation leftover from the Big Bang) to the amounts of hydrogen, helium, and heavier elements in our Universe. Because of this, even if the Universe has no end in space, it does have a beginning in time.

        When we look at these very distant galaxies, we’re actually looking back in time, rather than just looking at the far edge of the Universe. We can’t go back in time forever though! These galaxies live only (“only!”) about 1 billion years after the Big Bang, which is why they’re some of the very earliest galaxies in the Universe. You’re right though — building bigger telescopes will allow us to look at galaxies when they’re even younger, and so help us learn even more about how they formed. Hope this helps!

    • Real science might consider the red-shift theory may be in error, that freq shift is actually a microscopic loss of energy due only to extreme distances traveled. In other words, there is an additional term in all the equations that we are ignorant of !!! May Einstein please roll over … !!!

      • Hi Marv, I think what you’re describing is similar to the “tired light” idea, which Fritz Zwicky proposed. It’s a nice idea, but there isn’t any evidence in favour of it — even Zwicky eventually conceded that galaxy redshifts are due to the expansion of the Universe.

  2. Hi Roan, interesting read and really interesting results when compared to what Hubble obtained before.
    Given you are suggesting only two alternatives, I’m wondering what is your opinion on my suggestion that those far-away galaxies do not “belong” to our Universe, but to a previously present one? Imagine the Big Bang as a local event within a bigger playroom.
    Would this scenario contradict any current theory on this subject?
    Thanks amd hoping to hear back from you.

    • Hi Angel, this isn’t an idea I’m familiar with, but my understanding is that galaxies could only begin to form some time (tens of millions of years) after the Big Bang, which the results from this paper do not disprove. Galaxies are quite fragile structures really, and I don’t think one could survive in the extremely early Universe (very shortly after the Big Bang, when everything was hot and dense). Sorry I can’t be of more help, thanks for commenting though!

      • The big bang has never made much sense to me despite the overwhelming evidence. A big bounce makes more sense to me due the fact cycles happen constantly throughout nature and the universe. The fact information can’t be created or destroyed says to me it must have existed before hand for whatever reason. I think Hawking said it best that the universe exists because it has to. And Brian Green saying asking what came before this I a nonsensical question to which the answer is irrelevant

  3. The Big Bang explosion has a dynamics as per the coevolution theory of super massive black hole at the center of galaxy and stars.So the big bang occured as in a way very organised and never as random.

    • Galaxies only began to form a few hundred million years after the Big Bang, so although the Big Bang is indeed the origin of all the matter in the Universe, it isn’t really responsible for these kinds of galaxy properties. Things like the local environment of galaxies (i.e. whether they are in a galaxy cluster or not) play a much greater role in the evolution of galaxies.

      • What about multiuniverses and the cyclical nature of universes – so maybe there were universes before our one?

        • You’re absolutely right, this is a possibility! We currently don’t know whether we live in a “closed” Universe or not — a “closed” Universe is one where the average density of the Universe is high enough that eventually it will stop expanding, be overcome with gravity and collapse back to a single point, like it was at the Big Bang. In principle, this could then lead to another Big Bang, so maybe there were other Universes before ours!

      • Sorry, but looking at all these theories, the idea that this is all a simulation seems more reasonable to me.

        Where was this thing that exploded into making this universe? At a place called anti-space? What was it made of?

        Now with the leap in technology and information comes from JWST, it seems clearer that we just don’t know sh*t 🙂

        • There are plenty of people who would agree that we live in a simulation, it’s quite hard to prove whether or not we do! Regardless, it’s still important to try and understand the Universe we live in. If we are indeed in a simulation, our best chance of proving this is probably by learning more about the simulation we’re in!

      • The ‘universe’ in my thoughts “and my thoughts alone” as been here a lot longer than current thinking would have you believe.
        Imagine that we live in an helix, not a cone shaped universe, rather like that of DNA, but on a sum what larger scale.
        Depending on your exact location/ position within the helix, determines how far you can see, that is you only see back to a point. A singularity, an apparent starting point.
        But a very misguided thought / starting point.
        However this thought generates another problem, that is, there was no ‘BIG BANG’.
        There was no singularity.
        We only think that was the starting point because, like all things you look through that are twisting, you can only see to a point. That point is where the two edges merged together, which gives the appearance of a starting point, but is in fact only a corner, around which is hidden for all eternity the true vastness of the universe.
        Just my mad thoughts.
        We all have a view.
        Some go further than others.
        Some like mine are are doodles of the brain.
        Then there are the ones that are totally off the wall.
        But we each have an opinion, to be proved or dismissed.

        • “Doodles of the brain” are how a lot of theories start! Of course, the next important step is to use telescopes like JWST to collect evidence supporting or disproving these ideas, but there’s nothing wrong with a bit of imagination to get the ball rolling!

  4. Your big bang theory is looking more and more like a piece of Swiss cheese. But, you’ll carry on defending this utter nonsense because you’re indoctrinated. And quite silly

    • Hi Peter. Usually when we talk about the Big Bang theory, it’s part of the Lambda Cold Dark Matter (Lambda-CDM) model, which is a broader model of the Universe. Lambda-CDM explains a huge number of observations of our Universe, including the presence of the Cosmic Microwave Background (the leftover radiation from the Big Bang), the expansion of our Universe, and the formation of galaxies, galaxy clusters and cosmic filaments.

      Very few astronomers would claim that Lambda-CDM is perfect, because it isn’t: there are some things that it struggles to explain. However, this doesn’t mean that the whole theory is worthless. Instead, it just means that we need to make some tweaks to the theory, to make it more accurate. For example, the paper that I write about in this article shows that we need to make some changes to the theory, so that our models of galaxy mergers more closely resemble real-life mergers.

      This is how science advances — we start with a theory that is mostly accurate, and keep improving it to help us understand the Universe better and better. We use Lambda-CDM as our starting point is because, although it’s not perfect, it explains the Universe better than any other models that we have. I’m happy to discuss the evidence for Lambda-CDM and the Big Bang in more detail if you’d be interested. Hope this helps!

  5. What you thought about the space/time universe one second ago; well, it’s bigger!

  6. What about the revolutionary concept that “in the beginning God created the heavens and the earth“??? He could have created the universe to give the appearance of being older than it actually is… like someone setting up an aquarium with weathered rocks, and then adding the fish… The biggest problem with the Big Bang theory is that it still doesn’t explain where everything came from in the beginning!!! And even if it is expanding, it still had to start somewhere, with something… otherwise it contradicts another basic scientific law, the conservation of matter and energy, that nothing can be created or destroyed, but can only change form. It seems equally a leap of faith as believing that “God said, let there be light”…. Your thoughts? BTW, I’m a science teacher and yes, I believe in God! They are not mutually exclusive… Thank you for your insights into this fascinating topic!

    • It’s true that it’s not clear where the Big Bang came from, although answering this question is a very active area of research for a lot of scientists — the job of astronomers is to learn what we can from our Universe, based on facts and the evidence that we can measure. Whether there is anything beyond this is moving away from astronomy and more towards philosophy (an area in which I’m not really an expert, sorry!).

  7. Hi Roan, thanks for the super interesting article!

    Regarding the data in figure 2: it seems that there is a huge gap between Hubble and JWST predictions.
    I wonder how the data from Hubble was calculated if it did not have the required resolution at Z>2. If it had, shouldn’t the error-bars be much larger?

    There is only one sentence in the article that I disagree with 🙂
    You wrote “Tuning and re-tuning theories to fit new data is a normal part of the scientific process”. If we stick to the current scientific philosophy (e.g. by Popper) when a theory makes wrong predictions, then it must make new predictions with its new “tuned” version. Otherwise theories would be “tuned” forever…

    • Hi Tomer, this basically comes down to the fact that the Hubble results are not completely analogous to the JWST results, for a few reasons. One of the big ones is that Hubble and JWST look at different wavelengths of light — Hubble looks at visible light, while JWST looks at infrared light (because this light has been redshifted, these correspond to ultraviolet and optical light respectively, emitted from the galaxy).

      Also, these galaxies are very small, meaning that in Hubble many of them look like spheroid “blobs” when actually they are disks. This is a systematic error rather than a measurement error, so it wasn’t included in the original Hubble error bars.

      In answer to your second point — our current “theory” on the evolution of galaxies is really a combination of many different ideas. By tuning, I mean that we may need to change the relative contributions of different factors, because our current model is clearly not perfect (as this research shows). However, it still does a pretty good job at explaining galaxies, so we’re probably close! If we can tweak this theory to make it agree with this new observations, we’ll know that the tweaked version is even closer to a complete model of galaxy evolution.

      • Hi Roan,
        Thanks for your response.
        Do you know what are the relative fractions (of galactic types) in the current universe? or where can I find it….
        Thanks 🙂

        • I don’t know any detailed studies off the top of my head, but this paper by Schawinski et al. (https://arxiv.org/abs/1402.4814) looks at the star formation rates of galaxies between redshifts 0.02-0.05 (about 500 million years ago, which is approximately the present day!). They use data from the Galaxy Zoo project, which is an amazing project that asks large groups of amateur astronomers to classify galaxies, which anyone can get involved with. In Section 2.2 of that paper, they say that they find about twice as many “late-type” galaxies (dominated by a disky feature) compared to “early-types” (dominated by a blobby elliptical feature).

          This fraction is different in different parts of the Universe though. For example, galaxies in galaxy clusters are far more likely to be ellipticals than those outside of clusters. So, although it’s interesting to look at the total fraction of spiral/elliptical galaxies at a certain redshift, it doesn’t tell us everything!

  8. My biggest question was always, If the Universe is always expanding, and there is nothing bigger than the universe, what does it expand in to.

    • This is a good question, and one that there isn’t a clear answer to! To be clear though, the Universe isn’t really expanding into anything. It’s easy to imagine the Universe as being contained inside something bigger, and that it’s expanding to fill the extra space. However, that is not the case — the Universe is everything that exists, and so it is more accurate to say that the entire of existence is expanding, with everything moving further apart. It’s a tough thing to get your head around!

  9. Hi Roan,
    The most interesting part of the physical sciences is where improved observation proves that old theories are either incomplete or totally wrong, necessitating an overhaul or replacement with something unexpected. Hopefully the JWST is now doing just that to the entrenched theories of the BB and perhaps even general relativity itself.

    I am thinking of how the Michelson Morley experiment overturned the idea of an “ether”, and led to special relativity during that exciting time a century ago when SR, QM and the electromagnetic theories were being constructed. Will we finally see “dark energy” as just another “ether”, that evaporates under the spotlight of new data?

    Personally I have always liked the steady state theory of Hoyle et al, but evidence seemed to mount against it. Wouldn’t it be fantastic if it turned out to be correct after all. Imagine the turmoil! We already see a few cracks appearing in the long standing accepted cosmology: Planck found the universe to be flat, the inability to combine QM with GR, and now the Webbscope data.

    I have personally been involved with astronomy for decades, and IMAO astronomy has fallen into a pit of pop-culture innanity (I won’t go into the details here). At last we might have the opportunity to completely rewrite cosmology. I won’t bore anyone with my own speculations on cosmology, but I suspect that the universe is much, much simpler than it is cracked up to be.

    All we need now is a scope several times larger than the Webbscope! Unfortunately that won’t happen any time soon.

    • Hi John — you’re certainly right that it’s really exciting when we have to develop whole new areas of physics to explain new findings! I think we’re unlikely to see any convincing evidence for a steady-state model though, the evidence for a hot Big Bang is pretty overwhelming, and there is a huge amount of evidence showing how the properties of our Universe change over time. The results from this paper, while very interesting, still don’t give any evidence against the Big Bang.

      In fact, even the problems that you mentioned aren’t really problems (or at least, not new ones!). Planck’s measurements of a flat Universe are part of what motivates the study of cosmic inflation, which describes how the Universe expanded very shortly after the Big Bang. Inflation is a really active area of research, and also elegantly explains several other strange observations (like why regions of the Universe separated by vast distances have similar properties).

      Combining quantum mechanics with general relativity is a problem that has existed for almost as long as general relativity itself! While we know that GR is an extremely accurate approximation of the way our Universe works, we also know it isn’t perfect. One of its core predictions is that at the centres of black holes exist “singularities”, infinitely dense regions in space. However, this is forbidden by quantum mechanics.

      A theory successfully combining quantum mechanics and gravity would be one of the biggest jumps forward in the history of science, and I personally have no idea what such a theory would look like. I’m excited to find out though!


Leave a Reply