5

There are a finite number of currently observable galaxies due to the finite age of the universe and the speed of light. What fraction of these galaxies have we actually observed (by eye, telescope, etc.)?

christopherlovell
  • 3,160
  • 20
  • 31

2 Answers2

7

There are around 2 trillion galaxies in the currently observable universe according to the latest estimates, obtained by integrating theoretical galaxy stellar mass functions above $10^{6} M_{\odot}$ between $0 \leqslant z \leqslant 8$.

It's difficult to get a precise number for the total observed galaxies as the results from new surveys are being released all the time, and updated analysis of legacy data is revealing more galaxies. For example, the latest SDSS data release identified close to 200 million galaxies, whilst the ongoing Dark Energy Survey seeks to identify around 300 million galaxies. LSST, in Chile, will observe close to 20 billion galaxies, an order of magnitude greater than anything before it. However, all of these surveys operate up to relatively low redshifts. Probing to higher redshifts requires bigger, preferably space based telescopes, and much longer exposure times. As such, the high redshift universe has been poorly documented. Our understanding of galaxy formation at high redshift is not as complete as at low redshift, and so we may be under- or over-estimating the number of galaxies at this epoch.

A vague upper limit: we've observed less than 0.01% of all currently observable galaxies.

christopherlovell
  • 3,160
  • 20
  • 31
  • The number of galaxies in the observable Universe is increasing, not decreasing. The radius of the obs. Universe is increasing for two reasons: Firstly, the Universe is expanding. You're right that at a distance of the "edge", this expansion is at v > c, but this is no hindrance for observing stuff there. Secondly, even in comoving coordinates the radius is increasing, since light emitted from increasingly larger distances will have had the time to reach us. – pela Jan 30 '17 at 12:01
  • @pela "at a distance of the "edge", this expansion is at v > c, but this is no hindrance for observing stuff there" - I disagree. Galaxies at the edge of our observable horizon will be receding from us at a relative velocity > c, and will therefore drop out of our horizon at some point in the future (see http://physics.stackexchange.com/questions/5320/maximum-size-of-the-observable-universe). This is all dependent on current dark energy models. – christopherlovell Jan 30 '17 at 12:10
  • In comoving coordinates, the radius of the observable Universe will always increase, although it will asymptotically reach a maximum value (roughly 19-20 Gpc, depending on your cosmology). That means that the number of galaxies in the observable Universe will never decrease. Galaxies don't "drop out of our horizon", but they will become increasingly redshifted, and hence in practice impossible to detect. – pela Jan 30 '17 at 12:22
  • Superluminal velocities are no hindrance for being detected. For instance, the currently record holder for most distant galaxy, GN-z11, recedes at a speed more than twice the speed of light, and yet is easily detectable (well, maybe not "easily", but you can see it). – pela Jan 30 '17 at 12:26
  • …and the surface of last scattering of the CMB recedes at more than 3c, yet you can see, with your own eyes, (a small fraction of) the static in your old-scool TV set, caused by CMB photons. – pela Jan 30 '17 at 12:30
  • Also, you say there are 200 million SDSS galaxies. It's more like 2 million, right? I'm not an observer, but would Pan-STARRS perhaps contain more galaxies? – pela Jan 30 '17 at 13:05
  • @pela I see where you're coming from, but still a bit confused. "Superluminal velocities are no hindrance for being detected" of course not, but the light we see from GN-z11 was emitted ~ 12.5 billion years ago. The light it emits today will never be observable. "the surface of last scattering of the CMB recedes at more than 3c" yes, but the CMB is isotropic, not a discrete source close to the hubble horizon. – christopherlovell Jan 30 '17 at 14:42
  • @pela "Galaxies don't "drop out of our horizon", but they will become increasingly redshifted" agreed on extreme redshifts, but many sources suggest that the recession will lead to objects travelling through our hubble horizon (https://www.newscientist.com/article/mg21028061-400-empty-universe-cosmology-in-the-year-100-billion/ , https://en.wikipedia.org/wiki/List_of_cosmological_horizons#future_horizon , see Future Horizon) – christopherlovell Jan 30 '17 at 14:44
  • @pela "you say there are 200 million SDSS galaxies. It's more like 2 million, right?" - no, definitely 200 million added reference in the text – christopherlovell Jan 30 '17 at 14:46
  • 2
    Okay, I think we're discussing different horizons. You're right that light emitted today from, e.g., GN-z11 will never be observed. This is true for anything which today is more distant than ~5 Gpc. But you're asking about galaxies in the observable Universe which is something different. It consists of everything we will ever be able to see. We won't be able to see the light emitted today from a galaxy that today is, say, 6 Gpc away. But we can see the light that it emitted in the past, and hence, by definition, it lies inside the obs. Uni. – pela Jan 30 '17 at 15:09
  • The "future horizon" described in your Wiki link is what I referred to above: In comoving coordinates, the radius of the Obs. Uni. is ~14.5 Gpc. In increases monotinically, asymptotically reaching ~19 Gpc. That is, galaxies with d < 14.5 Gpc are in the obs. Uni; galaxies with 14.5 Gpc < d < 19 Gpc are outside, but will eventually be inside; galaxies with d > 19 Gpc will never be inside. – pela Jan 30 '17 at 15:09
  • In physical coordinates, the radius of the obs. Uni. increases without bounds. But this doesn't increase the number of galaxies we will be able to see, as they'e carried away too. In fact, in practice it decreases the number, since their redshift asymptotically goes to infinity. – pela Jan 30 '17 at 15:09
  • I guess we should go to chat if we want to discuss further… – pela Jan 30 '17 at 15:09
  • @pela agreed, or we could settle this in a separate question? I now doubt my explanation, but I'm still not completely convinced by your reasoning (sorry) – christopherlovell Jan 31 '17 at 14:59
  • @pela Given the disagreement I'll remove the statement on time dependence of observable galaxies, and restrict the answer to observed/observable galaxies today – christopherlovell Jan 31 '17 at 14:59
  • See my answer here which may help. – pela Jan 31 '17 at 16:57
  • Could you give a proper provenance and description for the number of 2 trillion galaxies in the observable universe, since this is critical for your answer. My guess would be that this is a speculative number based on the integration of some assumed galaxy luminosity function down to some limit. It therefore doesn't include fainter galaxies, where the sample is not even complete in our local neighbourhood. So you are correct; 0.01% is very much an upper limit and the answer is approximately zero, as per my comment. – ProfRob Apr 02 '17 at 10:19
  • @RobJeffries added link to article and explanation of the derivation. It's an empirical estimate based on GSMF fits up to $z = 8$. They use a lower stellar mass limit of $10^{6} M_{\odot}$, which covers typical dwarf galaxy masses, but the low mass end is pretty much guesswork at $z = 8$. – christopherlovell Apr 02 '17 at 15:42
5

The problem is that we don't really know how many small and diffuse galaxies there are. Even in our own cosmic backyard, the local group, we are still discovering new galaxies. Since these dwarf spheroidal galaxies are by far the most common, we have essentially observed a negligible fraction of all galaxies in the observable universe, and will never be able do increase this significantly.

Walter
  • 5,546
  • 1
  • 19
  • 28