Galaxy Count: How Many in the Visible Universe?

[TalonD]

rough estimate, how many galaxies in the visible portion of the universe? Just curious.

 

[Wallace]

Edit: Sorry, I did post an incomplete answer, but I stuffed up somewhere. I eneded up calculating there are about 10 thousand galaxies. Which is just a little bit too small...!

Anyone else want to give this a shot? I will try and work out where I went wrong?

 

[nicksauce]

From Wikipedia: 'There are probably more than 100 billion galaxies in the observable universe", with the reference: http://astronomy.swin.edu.au/~gmackie/billions.html

But we can do an order of magnitude calculation to estimate this.
Assumptions:
1) All the baryonic matter in the universe is in galaxies, in the form of stars
2) The universe is flat with $\Omega_Bh^2 = 0.02$
3) Each galaxy has 100 billion stars, averaging 1 solar mass
4) The size of the observable universe is about 4 * 10^32 ly^3 (http://en.wikipedia.org/wiki/Observable_universe)

This gives about 6 * 10^11 galaxies, close to the 100 billion quoted by Wikipedia.

 

[russ_watters]

You can also reach a similar estimate by taking the Hubble Ultra Deep Field (which contains around 10,000 galaxies) and extrapolating it across the whole sky.

 

[negitron]

Estimating the exact number of galaxies in the observable universe is a challenging task due to the vastness of space and the limitations of our current technology. However, astronomers have made significant efforts to estimate this number. The most recent estimates suggest that there are approximately 2 trillion (2 x 10^12) galaxies in the observable universe.

This estimate is based on observations made by space telescopes like the Hubble Space Telescope and deep-sky surveys. These observations reveal that the universe is teeming with galaxies, each containing billions to trillions of stars. The observable universe represents only a fraction of the entire universe, and it continues to expand, so the actual number of galaxies, including those beyond our observational reach, may be even larger.

It's important to note that our understanding of the universe is continually evolving as technology advances and our observations improve. Future discoveries and advancements in cosmology may lead to more accurate estimates of the number of galaxies in the universe.

 

[Chronos]

Agree with nick, over 100 billion galaxies, so far. The universe is really gigantic, relatively speaking.

 

[TalonD]

You can also reach a similar estimate by taking the Hubble Ultra Deep Field (which contains around 10,000 galaxies) and extrapolating it across the whole sky.

Does that mean that Hubble can see all the way to the earliest forming galaxies ?

 

[Wallace]

Hubble, being the best telescope we have ever had for this sort of thing, has seen the most distant galaxies ever observed (measured by their redshift). There is a fair bit of uncertainty over exactly what redshift the furthest Hubble Ultra Deep Field galaxies are, since all you really get is a hazy blob seen in a few different filters (e.g. in different colours) from which you make an informed guess about the redshift, based on what colours you expect galaxies to be in their rest frame.

I believe (if anyone could provide a reference that would be good) that Hubble has seen galaxies as distant (in terms of redshift) that it might be expected to be able to. Therefore an even more powerful telescope could in principle see even more distant galaxies, if indeed galaxies existed at such high redshifts, which is not a sure thing.

In terms of a ball park estimate of the total number of galaxies in the observable universe though, these extra ones that we can't yet see (if they exist) won't change the figures people have quoted in this thread by much.

 

[Blenton]

One thing, when we look at the Hubble deep field, how do we know there arent any galaxies behind the galaxies we see there. Id assume that light wouldn't pass through.

 

[Wallace]

To some extent that it is true, although there are enough gaps through which we can see nothing in the way. On the other hand, if the separation in redshifts between the galaxies fits a certain criteria, it is actually better if the further one is behind the first, since the gravity of the foreground galaxy can act as a graviational lens, in fact allowing us to see the more distant galaxy with a greater intensity than otherwise. Some of the most distant objects seen have been discovered this way.

 

[negitron]

...all you really get is a hazy blob seen in a few different filters (e.g. in different colours) from which you make an informed guess about the redshift, based on what colours you expect galaxies to be in their rest frame.

Huh? I do not think you understand the process here. Redshift measurement doesn't involve guesswork except to the extent that it's difficult to get readable spectra from such distant, faint objects. Once a clean spectrum has been obtained, however, redshift can be measured precisely because it doesn't involve guessing about colors, it looks for specific signature spectral emission and/or absorption lines, such as the ones for hydrogen and measures exactly how far it has been shifted towards the red (or blue, as the case may be). The color of the galaxy is irrelevant.

 

[Wallace]

Huh? I do not think you understand the process here. Redshift measurement doesn't involve guesswork except to the extent that it's difficult to get readable spectra from such distant, faint objects. Once a clean spectrum has been obtained, however, redshift can be measured precisely because it doesn't involve guessing about colors, it looks for specific signature spectral emission and/or absorption lines, such as the ones for hydrogen and measures exactly how far it has been shifted towards the red (or blue, as the case may be). The color of the galaxy is irrelevant.

Hi Negitron, what you are saying is true in some cases, so it is good to have it pointed out. There are indeed two main ways to measure redshift, spectroscopic and photometric. Spectroscopic is very very accurate (if you get a signal at all, you know the redshift to 2 or 3 decimal places at least), but is expensive in terms of telescope time, since making a spectrum needs many more photons than an image. Photometric redshift measurements on the other hand is a technique to estimate the redshift based on in image in only a few colour filters, based on the knowledge of the major features in the spectrum. Photometric redshifts are used in, for instance, weak lensing surveys to get many many redshifts of background galaxies near gravitational lenses. While this is less accurate than spectroscopic, it is impractical (would take more telescope time than is available) to get spectroscopic redshifts.

Specifically in the case of the HDF and the HUDF, photmetric redshifts are the only ones possible since these surveys were done using the Hubble wide field camera in a number of colour bands. This means that photmetric, rather than spectroscopic, redshifts must be used. It simply isn't possible using current technology to get spectra of such distant objects.

 

[negitron]

Ah, it wasn't immediately clear to me that you were speaking of photometric analysis rather than spectroscopic. If I'd given it some thought, it would have occurred to me that, of course, Hubble can't do spectroscopic analysis of such distant objects. Why? not because the technology doesn't exist, it's simply that at the time Hubble was deployed, it wasn't known that there were such distant objects!

 

[Wallace]

Hubble has a spectrograph, it did get broken, but has been replaced in a recent servicing mission. Hubble spectra are pretty crappy though, because the Hubble mirror is so small (only ~2m). Ground based telescopes are actually better than Hubble for spectra because the distortion effects of the atmosphere are not a problem, so it is a case of bigger is better. The highest redshift spectra have been obtained from the ground, not from space.

The reason Hubble has a spectrograph is to get spectra in the UV, which can't be taken from the ground because of the ozone layer. This generally means objects either in our own galaxy or relatively nearby, since UV emission from high redshift galaxies gets shifted into the visible (or infrared in extreme cases).

 

[russ_watters]

Hubble, being the best telescope we have ever had for this sort of thing, has seen the most distant galaxies ever observed (measured by their redshift).

I was under the impression there was a different type of telescope (a radio telescope maybe?) that could see galaxies further away, but can't remember where I heard that. In any case, yeah, Hubble probably gets us on the right order of magnitude.

 

[TalonD]

Correct me if I'm wrong, and I very well may be considering I'm just a layperson. But wouldn't the most distant galaxies we can see be limited by the surface of last scattering, that is, the microwave background radiation. so the radius of the visible universe is around 45 billion light years? So is hubble's resolution high enough to image galaxies or protogalaxies that far away?

 

[matt.o]

Hubble has a spectrograph, it did get broken, but has been replaced in a recent servicing mission. Hubble spectra are pretty crappy though, because the Hubble mirror is so small (only ~2m).

Yes, the http://www.stsci.edu/hst/cos" which will hopefully be taking spectra within the next few months.

Ground based telescopes are actually better than Hubble for spectra because the distortion effects of the atmosphere are not a problem, so it is a case of bigger is better. The highest redshift spectra have been obtained from the ground, not from space.

Speaking from experience (mainly at the AAT...), bad seeing is most definitely a problem for collecting spectra, in particular for higher redshift objects, where bad seeing will mean you'll have to significantly increase your exposure time!

 

[Wallace]

Ah okay. I was comparing Hubble spectra to Keck spectra that I worked on years ago, the Keck ones were waaaay better than Hubble. So seeing is still a problem for getting spectra (which makes sense) but not so bad such that a 10m ground based telescope still beats (by a fair way) a 2m space telescope, which is not the case for imaging?

 

[matt.o]

Ah okay. I was comparing Hubble spectra to Keck spectra that I worked on years ago, the Keck ones were waaaay better than Hubble. So seeing is still a problem for getting spectra (which makes sense) but not so bad such that a 10m ground based telescope still beats (by a fair way) a 2m space telescope, which is not the case for imaging?

Yes, Keck will definitely do a better job than Hubble as it has way more collection area, as you correctly pointed out!

 

[TalonD]

From Wikipedia: 'There are probably more than 100 billion galaxies in the observable universe", with the reference: http://astronomy.swin.edu.au/~gmackie/billions.html

But we can do an order of magnitude calculation to estimate this.
Assumptions:
1) All the baryonic matter in the universe is in galaxies, in the form of stars
2) The universe is flat with $\Omega_Bh^2 = 0.02$
3) Each galaxy has 100 billion stars, averaging 1 solar mass
4) The size of the observable universe is about 4 * 10^32 ly^3 (http://en.wikipedia.org/wiki/Observable_universe)

This gives about 6 * 10^11 galaxies, close to the 100 billion quoted by Wikipedia.

6*10^11 is 600 trillion, that's not quite the same as 100 billion

 

[negitron]

6*10^11 is 600 trillion, that's not quite the same as 100 billion

Eh? 6E11 = 600,000,000,000 = 600 billion.

 

[wil3]

I believe Carl Sagan would say "Billions and billions"

 

[TalonD]

Eh? 6E11 = 600,000,000,000 = 600 billion.

did I get a decimal or 2 in the wrong place? Yeah I did didn't I!
ok, well, 600b is still five hundred billion more than 100 billion. I wouldn't say that was close.


heh billions and billions..

 

[ideasrule]

Correct me if I'm wrong, and I very well may be considering I'm just a layperson. But wouldn't the most distant galaxies we can see be limited by the surface of last scattering, that is, the microwave background radiation. so the radius of the visible universe is around 45 billion light years? So is hubble's resolution high enough to image galaxies or protogalaxies that far away?

The microwave background radiation was emitted when the universe was only 400 000 years old. At that time, the baryonic matter of the universe had just begun to cool down into a gas from a plasma. There were no stars, let alone galaxies. The first galaxies formed a few hundred million years after the Big Bang, so none of them are "behind" the surface of last scattering.

As for Hubble's imaging capabilities, it certainly can't see every galaxy there is to see. That's part of the reason Herschel (see http://www.space.com/scienceastronomy/090512-herschel-sidebar.html) was launched: to see some of the earliest galaxies.

 

[Wallace]

600b is still five hundred billion more than 100 billion. I wouldn't say that was close.

Given the very simplistic assumptions used, a factor of six is neither here nor there. In astronomy you have to re-calibrate your idea of what is accurate. Often just having the right number of zero's in your answer is a pretty good achievement.

 

[Chronos]

I believe the 100 billion galaxies estimate came from the HDF [hubble deep field] survey [which covered a miniscule slice of the sky]. As Wallace noted, any such estimate accurate to the nearest zero is usually considered very good in cosmology.

 

[TalonD]

The microwave background radiation was emitted when the universe was only 400 000 years old. At that time, the baryonic matter of the universe had just begun to cool down into a gas from a plasma. There were no stars, let alone galaxies. The first galaxies formed a few hundred million years after the Big Bang, so none of them are "behind" the surface of last scattering.

As for Hubble's imaging capabilities, it certainly can't see every galaxy there is to see. That's part of the reason Herschel (see http://www.space.com/scienceastronomy/090512-herschel-sidebar.html) was launched: to see some of the earliest galaxies.

correct in a sense, however the matter that emitted the radiation that we see as the microwave background, emitted that light some 13.7 billion years ago and has since condensed into stars and galaxies even though we can't see them yet. And if there were some critter on a planet way over there looking back in our direction would it see our galaxy? no it would see the microwave background radiation from the matter that eventually collapsed to become us. So I would think if Hubble has good enough resolution then wouldn't it be able to see galaxies in the early stages of forming? Sometime after the emission of the background that is.

ah here it is...
this galaxy is so far away its light has been stretched out of the visible spectrum
A1689-zD1
http://hubblesite.org/newscenter/archive/releases/2008/08/image/a/

 

[TalonD]

I believe the 100 billion galaxies estimate came from the HDF [hubble deep field] survey [which covered a miniscule slice of the sky]. As Wallace noted, any such estimate accurate to the nearest zero is usually considered very good in cosmology.

I see your point. and I did ask for a rough idea, just out of curiosity.