≡ Menu

Cosmology: Shelter from the Storm

I had thought while the power was out this past week that I would like to write about cosmology when it came back. That’s because there’s nothing like a prolonged power outage to adjust your perspective. The big picture beckons. In my case, it was thinking about how trivial being out of power was compared to those who had lost so much more in the wake of the recent hurricane.

So thinking about the cosmos became my shelter from the storm. I appreciated the emails from so many of you, but aside from a major chunk of tree that landed on the roof, we did just fine. In fact, it was deeply moving to see people from the neighborhood — some I knew, some I only recognized — turn up to get up on the roof and move that tree. I’m always reminded to do more for the people around me when I see something like this, and apprehensive that my resolution to do so all too often gets put aside as normal life returns.

The Universe We Can See

Reading by candlelight really is wonderful, and I ask myself why I don’t do it more often. There is something about that soft, flickering light on a well-printed page. And I found reading about cosmology by candlelight was especially pleasing, a way of connecting to a past way of life that had its own conceptions about the cosmos. One thing I read was my notes from work just received when the power went, a paper on galactic structure I fortunately printed just in time.

The conclusion of this one is straightforward: There are at least 10 times as many galaxies in the observable universe as previously thought. Christopher Conselice (University of Nottingham, UK) led an international team that produced this result. I was reading about it in a room full of candles and thinking about room after room of candles stretching out into infinity.

That homely thought weds the prosaic and the vast, sort of the way Douglas Adams did in his famous line on the size of the universe in The Hitchhiker’s Guide to the Galaxy:

“Space is big. You just won’t believe how vastly, hugely, mind-bogglingly big it is. I mean, you may think it’s a long way down the road to the chemist’s, but that’s just peanuts to space.”

I’m sure you’re familiar with the line, but it still raises a chuckle. And this team’s findings cause yet another shift in perspective. Thus Conselice:

“It boggles the mind that over 90% of the galaxies in the Universe have yet to be studied. Who knows what interesting properties we will find when we observe these galaxies with the next generation of telescopes.”


Image: The GOODS South Field (Great Observatories Origins Deep Survey). GOODS draws on observations from the Spitzer, Hubble and Chandra observatories as well as the European Space Agency’s Herschel and XMM-Newton, along with ground-based facilities, to survey the distant universe across the electromagnetic spectrum. Credit: GOODS/Conselice et al.

Before this, we relied on such measures as the Hubble Deep Field images from the 1990s, which helped us arrive at an estimate of between 100 and 200 billion galaxies in the observable universe. Conselice and team studied the matter using Hubble imagery as well as other published data to create a 3D view, estimating the number of galaxies at various times in the history of the cosmos. Their mathematical models pointed to the result, that we have not been able to see most of the galaxies that are out there in our cosmological horizon.

Moreover, the work offers an interesting perspective on galaxies throughout time, one showing how their numbers have changed as the universe evolved. The paper explains that the total number density of galaxies in the universe declines with time:

[The] total number density… declines by a factor of 10 within the first 2 Gyr of the universe’s history, and a further reduction at later times. This decline may further level off between redshifts of z = 1 and z = 2. The star formation rate during this time is also very high for all galaxies, which should in principle bring galaxies which were below our stellar mass limit into our sample at later times. This would naturally increase the number of galaxies over time, but we see the opposite. This is likely due to merging and/or accretion of galaxies when they fall into clusters which are later destroyed through tidal effects, as no other method can reduce the number of galaxies above a given mass threshold.

So what is known as the ‘top-down formation of structure’ in the universe is supported by this work. Most of the galaxies in the first few billion years of the universe were similar in mass to the satellite galaxies we see surrounding the Milky Way. Galactic mergers reduce the total number. And although Conselice and colleagues show us that there are so many galaxies that almost every point in the sky contains part of one, most of these galaxies are invisible not just to the human eye but to our best telescopes.

If that rings a bell, it’s because we’re homing in on Olbers’ paradox, the idea that the darkness of the night sky is in contradiction to the assumption of an infinite universe filled with stars (based on the work of the German astronomer Heinrich Wilhelm Matthias Olbers, 1758-1840). Why is the night sky as dark as it is? I won’t get into the details on dust and gas absorption of light that the paper offers, but will simply quote the finding from the Conselice paper:

It would… appear that the solution to the strict interpretation of Olbers’ paradox, as an optical light detection problem, is a combination of nearly all possible solutions – redshifting effects, the finite age and size of the universe, and through absorption.

The paper is Conselice et al., “The Evolution of Galaxy Number Density at z < 8 and its Implications,” to be published in the Astrophysical Journal. Preprint here.


Comments on this entry are closed.

  • David October 14, 2016, 10:04

    Glad you are back. Did you get a look at the lightless night sky. Its better here where its just tornados we tend to only be without piwer overnight. Always wondered about Oblers Paradox

  • Robin Datta October 14, 2016, 10:36

    Glad you were spared the worst of the storm, and you sagely contemplated a few of the bigger questions about our universe.

    Of course, venturing into the “ten thousand things” excludes a quest for limits because of its infinitude: the “ten thousand things” is a Taoist & Buddhist reference to the variety, complexity and non-finiteness of all that is percievable and concievable about the phenomenal world.

  • JonW October 14, 2016, 11:25

    Wonderful that there’s such a large number of galaxies about to become observable… but simultaneously a little sad that there is now so much more of the observable universe that is destined to remain forever unreachable, whatever new physics our descendants develop. No doubt all these “new” galaxies have redshift > 1.5 and hence are causally disconnected from us, currently unreachable even at light speed.

  • Shaun October 14, 2016, 12:19

    Paul, you are getting positively poetical!

    • Paul Gilster October 14, 2016, 13:40

      Thanks! I do love going after the occasional poetic effect… ;-)

  • DJ Kaplan October 14, 2016, 12:31

    Welcome back! Interesting subject.

    When we say that the total number density of galaxies in the universe declines with time, does that mean that the farthest galaxies we can see (assuming that these are the oldest) are less dense than the ones that are closest to us?

    • Paul Gilster October 14, 2016, 13:39

      The way I read this paper, those oldest visible galaxies would be smaller and less crowded with stars than galaxies much closer to us.

  • Harry R Ray October 14, 2016, 13:15

    Is it reasonable to infer that there are ALSO ten times as many STARS in the observable universe(currently estimated at 4 SEXTILLION)as previously thought?

  • Montie October 14, 2016, 15:00

    What, if anything, does this have to do with dark matter or dark energy?

    • Mark Zambelli October 20, 2016, 9:50

      Nothing. Those are heavily constrained by the map of the CMB at 380,000yrs after the HSK, (horrendous space kablooie). Then we have all the structure that we can see in the observable Universe today, going back as far as z=8-ish, thanks to those handful of red ‘dots’ in the Deep Field.

      This newly reported twist on the number of galaxies we should be acknowledging concerns the ‘stuff’ in between these two eras (between the imprints of seed-structure in the CMB, and our models of how early large galaxies grew) In terms of building-blocks, we’ve learned that more of the smaller galaxies are needed to explain the number of observable galaxies we see today so their existence is implied. This doesn’t mean 10 times more stuff just came into being… it just means our estimates and inferences so far have fallen short and we now know something more accurately.

  • David W Cummings October 14, 2016, 16:19

    Beautiful post, Paul. Seriously. I’m going to stop off at a library, take out a book on cosmology, and go home and read it by candle light.

    • Paul Gilster October 14, 2016, 17:06

      You won’t be disappointed! Enjoy the experience.

  • ljk October 14, 2016, 16:38

    About Those 2 Trillion New Galaxies . . .

    By: Alan MacRobert | October 13, 2016

    Maybe you saw today’s news that the universe contains “10 times as many galaxies as astronomers previously thought.” Well, there’s 10 times less to that announcement than meets the eye. But the real news is interesting enough.

    NASA, the European Space Agency, and others issued press releases this morning announcing a new analysis of multi-instrument censuses of the Hubble Ultra Deep Field. A team led by Christopher Conselice (University of Nottingham, UK) sorted galaxies by their brightnesses and estimated redshifts to create a 3D model of which ones are at different distances and hence cosmic ages. Their study provides a better look at galaxy evolution over time, and estimates how many galaxies are too faint and far to see.

    Full article here:


    To quote:

    Up to now, astronomers usually said we know of about 200 billion galaxies in the observable universe (meaning out to our event horizon, a look-back time of 13.8 billion years). Now the number can be said to be about 2 trillion, with the caveat that this estimate doesn’t go back a full 13.8 billion years, it’s 600 million years short. (Not many galaxies could have formed before then.)

    The only reason the number is 10 times bigger now is that you can legitimately include more of those littlest early building blocks; they’re no longer so theoretical. The total amount of stuff — stars and gas — hasn’t changed.

    So no, we do not “also have to update the number of stars in the observable universe, which now numbers around 700 sextillion,” as some uninformed science writers are saying. That’s what they get for taking press-release hype literally.

    • David W Cummings October 15, 2016, 8:55

      As soon as I first read the story about “ten times more galaxies” I noticed there wasn’t a breath of a mention about a total increase in the amount of stuff in the universe (or increase in our estimation of the amount of stuff), and in every subsequent story that glaring lack of mention was repeated with religious-dogma correctness. The total number of writers in the media who can actually think for themselves is extremely low. (They are mostly all here, in this blog.)

      Anyway, regarding all that stuff (the estimate of which remains unchanged), all those 700 sextillion stars, it’s worth remembering that there is a big question of why they exist at all, since the big bang should have created matter and anti-matter in equal amounts. But it didn’t. The reasons put forward why it didn’t are varied but the estimate of how much stuff survived the matter-antiMatter annhilation is ONE PARTICLE IN A BILLION.

      In other words, there was an imbalance of matter and anti-matter at the big bang that amounted to 500,000,001 matter particles for every 500,000,000 anti-matter particles.

      That means that those 700 sextillion stars represent one one-billionth of all the stuff that was initially created in the big bang.


      • hiro October 17, 2016, 20:28

        An obvious remark: antiparticles travel back in time due to the opposite time arrow hence the process of trapping antiparticles could be understood as stopping them from going to the past. It’s possible for antimatter that exists in the future moving back through time but being stopped at the present by antimatter trapping devices; normally it’s like walking pass some strangers in very dense fog.

        • Mark Zambelli October 20, 2016, 9:19

          I still love the mischievous idea (without checking, possibly by Paul Davies…?) as to why all electrons/positrons are identical… there’s only one! A single electron was created at the BB and it travelled forward in time until the heat death where it became a positron with it’s time-arrow flipped and so travelled back to the BB, passing itself when it was the electron. There it became an electron and moved forward again but this time in a Universe that already had itself in it. Repeat these ‘bounces’ between beginning and end about 10^80 times and we see a universe with that many copies of the single electron. Cute :D

          • Paul Gilster October 20, 2016, 11:09

            This is a Feynman notion, if memory serves. I love it.

            • Mark Zambelli October 24, 2016, 11:19

              (Thanks Paul… I now recall it was indeed Feynman and that I merely learned of it through Paul Davies. Yet another doozie from the man himself, Feynman!)

          • hiro October 26, 2016, 1:35

            The BB created 2 universes: this one and the other one which has the opposite time arrow happens to contains anti-matter. Time travel could be understood as the process of transforming matter->antimatter->matter to a desired period + location since the location of Earth in this month is different from location of the Earth in last October. Anyway, sending positrons to the past is easy but sending them in Morse code is highly nontrivial (by using the properties of spin quantum #).

    • David W Cummings October 15, 2016, 9:07

      The real story behind the “ten times more galaxies” is both more prosaic and more exciting (at least from a human perspective):

      The real story is that the increase in telescope technology of all kinds is allowing us to get more and more detail from the earlier epochs of the universe.

      And that is a very big story indeed, and continuing to unfold. I can’t wait for the JWST!

  • Stevo Darkly October 14, 2016, 20:11

    Hi, Paul. Glad to have you back!

    Just thought someone should mention this: It looks like maybe you have an unclosed italics tag at the end of this post (where you cite the Astrophysical Journal reprint) and it’s turning the whole rest of the blog into italics.

    • Paul Gilster October 14, 2016, 21:14

      A thousand thanks — I did see the problem but hadn’t gotten it fixed until I went searching for the italics tag. Much appreciated!

  • Phil October 14, 2016, 22:21

    I wonder sometimes if we are too ‘galaxy-centric’. I’m sure there have been some estimates regarding the number of stars (and presumably planets) that are wandering in the intergalactic voids, presumably as a result of ejection events, but maybe also as a result of galactic interactions with clouds of isolated cold gas? Either way these bodies must number in their trillions. I wonder how much ‘missing mass’ is out there in the lonely dark?


  • George King October 15, 2016, 19:27

    “Beauty walks a razor’s edge, someday I’ll make it mine
    If I could only turn back the clock to when God and her
    were born,
    Come in, she said
    I’ll give ya’ shelter from the storm.”

    Bob Dylan
    “Shelter From the Storm”
    Blood on the Tracks

    • Paul Gilster October 15, 2016, 20:33

      Absolutely right, George. It’s always been a favorite of mine, and was indeed the reference I was making.

      • ljk October 17, 2016, 9:17

        That Dylan guy should get a Nobel Prize for his writing….

  • kzb October 17, 2016, 8:29

    Interstellar absorption cannot be an explanation of Olber’s paradox.

    Also there is a big problem with the surface brightness of these early galaxies. They are (apparently) 1000 times brighter than any local galaxy. Worse, there is apparently a maximum surface brightness in the local universe which no part of any galaxy exceeds. Yet these early galaxies exceed it by 1000-fold. Google the Galaxy Ancestor Problem.

Next post:

Previous post: