Number of galaxies in the Universe

[davidge]

This may sound like a noob question, even more if it's asked by a undergraduate physics student, but here it's:

How do we know about the existence of 100 billion galaxies out there (even that it's an average value)? I mean, how do we know about these exact number? Why not, say, 200 billion or 50 billion?

Is there a concrete way to obtain the value out of the Relativity Theory? If not, what theory should we use to get 100 billion?

 

[phinds]

I'm not aware of any "theory" that arrives at that number, it's just an estimate based on empirical evidence, much like how a crowd of 1,000,000+ is estimated. You cut out a small average section, count it carefully, and extrapolate to the total area (volume in the case of galaxies). I've seen recent estimates that dramatically up the size of the estimate for the observable universe (which, by the way, is what you are talking about, NOT "the universe", which may be infinite in extent and in any case is WAY bigger than the observable universe) so the estimate is apparently not very precise.

 

[phyzguy]

As phinds said, it is just an estimate. Where did you get the idea that it was an exact number? I would characterize it as an order-of-magnitude estimate, meaning that the number of galaxies in the observable universe is probably somewhere between 10^10 and 10^12. There's also some ambiguity in just how large an assembly of stars needs to be in order to be counted as a galaxy.

 

[davidge]

Where did you get the idea that it was an exact number?

I mean a exact value on an average estimate.

probably somewhere between 10^10 and 10^12

How do you get that?

 

[phyzguy]

I mean a exact value on an average estimate.

What does an "exact value on an average estimate" mean? If it's an estimate, how can it be exact? If you asked for an estimate of the number of grains of sand on Waikiki beach, would you expect an "exact" value?

How do you get that?

I was just explaining what an "order of magnitude estimate" of 10^11 galaxies means. As phinds explained, the actual estimate is obtained by counting galaxies in some small region of space, then extrapolating to the size of the observable universe.

 

[davidge]

Hmm, ok.

But why is the result completely empirical? It seems a bit bizarre that we can't get the value from any of our theories. It's as if we were 10,000 years in the past, on a time when people knew everything only through observations.

 

[phinds]

Hmm, ok.

But why is the result completely empirical? It seems a bit bizarre that we can't get the value from any of our theories. It's as if we were 10,000 years in the past, on a time when people knew everything only through observations.

Well, YOU come up with a theory that explains it. I see no reason why there should be a theory that explains it any more than a theory that explains how many grains of sand there are on planet Earth.

 

[Grinkle]

In some manner or maybe in various manners, physicists model the early universe and as the density decreases they can describe different epochs (eg https://en.wikipedia.org/wiki/Chronology_of_the_universe ) based on the model. It might be that predicting the number of stars / galaxies that will form on average per unit volume is within the scope of these models - does anyone know if it is?

 

[PeterDonis]

It seems a bit bizarre that we can't get the value from any of our theories.

Why? To run with the analoguy @phyzguy made, would you expect a theory of physics to give you the exact value for the number of grains of sand on Waikiki beach? Isn't it obvious that that number is just the accidental result of a bunch of random factors combining over time? Why would you expect the number of galaxies in the universe to be any different?

 

[phyzguy]

In some manner or maybe in various manners, physicists model the early universe and as the density decreases they can describe different epochs (eg https://en.wikipedia.org/wiki/Chronology_of_the_universe ) based on the model. It might be that predicting the number of stars / galaxies that will form on average per unit volume is within the scope of these models - does anyone know if it is?

Well, there are detailed simulations of the evolution of the universe from early times up to the present day. Two examples are the Illustris simulation and the Eagle simulation. These are both fascinating studies by the way. Both of these give galaxy distributions which closely match what we observe. So I guess you could consider this "theory" predicting the number of galaxies in the observable universe. But these simulations have a large number of adjustable models which are tuned to give the right answers, so it is not theory in the sense of first principles theory.

 

[kimbyd]

This may sound like a noob question, even more if it's asked by a undergraduate physics student, but here it's:

How do we know about the existence of 100 billion galaxies out there (even that it's an average value)? I mean, how do we know about these exact number? Why not, say, 200 billion or 50 billion?

Is there a concrete way to obtain the value out of the Relativity Theory? If not, what theory should we use to get 100 billion?

As far as I know, the 100 billion number came from the Hubble Ultra Deep Field:
https://www.spacetelescope.org/images/heic0611b/

This image has nearly 10,000 galaxies in it. Divide the total area of the sky by the area of this patch, and then multiply that by the 10,000 galaxies in this patch and you get about 100 billion.

That said, there are probably many galaxies in this image that you cannot see (because they're too dim), so chances are this is a significant underestimate. By how much, I do not know. There could easily be 200 billion galaxies out there.

 

[davidge]

@kimbyd, good, thanks.

 

[laurenrayjohn]

Great replay @kimbyd. I totally agree with your point of view!

 

[newjerseyrunner]

When making claims like that, 100 billion and 200 billion are essentially the same thing. It's understood that because of the methodology, any answer is only going to be an approximate answer, so the order of magnitude is what's interesting, not the first digits. If someone tells me that there are 100 billion galaxies without mentioning a standard deviation, I take that to mean anywhere between 50 billion and 500 billion.

 

[kimbyd]

When making claims like that, 100 billion and 200 billion are essentially the same thing. It's understood that because of the methodology, any answer is only going to be an approximate answer, so the order of magnitude is what's interesting, not the first digits. If someone tells me that there are 100 billion galaxies without mentioning a standard deviation, I take that to mean anywhere between 50 billion and 500 billion.

Nit: order of magnitude estimates are more accurately done if the cutoff is $\sqrt{10}$, which is about 3.2, so that 100 billion order of magnitude is anywhere from 32 billion to 320 billion (roughly).

That aside, this is better understood as a lower bound than an order-of-magnitude estimate. Depending upon how small a cluster of stars you want to consider to be a galaxy (e.g. would the large and small Magellanic clouds count?), the total number of galaxies might be in the trillions to tens of trillions.

 

[Thuring]

I understand someone did a very recent estimate using some sort of new method that is apparently well accepted: 2 TRILLION galaxies within the visible universe. It used to be about 200 million. If you look it up, be sure it's a very recent report.

 

[Thuring]

I understand someone did a very recent estimate using some sort of new method that is apparently well accepted: 2 TRILLION galaxies within the visible universe. It used to be about 200 million. If you look it up, be sure it's a very recent report.

I ment "200 billion".

 

[kimbyd]

I think my post #11 above also answers this question. If the team making the claim of more galaxies, they may just be using an observational method that goes quite a bit deeper than Hubble's Ultra Deep Field.

 

[davidge]

@kimbyd, there's a little problem with your explanation in post #11. That's it the given photo was taken between 2003 and 2004. The value 100 billion I referred to was mentioned in the Cosmos Series in the early 80's. So there must be a source for that total number other than the Hubble's Ultra Deep Field.

 

[phinds]

I understand someone did a very recent estimate using some sort of new method that is apparently well accepted: 2 TRILLION galaxies within the visible universe. It used to be about 200 billion. If you look it up, be sure it's a very recent report.

This NASA site doesn't give a number but "10 Times More Galaxies Than Previously Thought" would imply 1 to 2 trillion
https://www.nasa.gov/feature/goddar...0-times-more-galaxies-than-previously-thought

 

[kimbyd]

@kimbyd, there's a little problem with your explanation in post #11. That's it the given photo was taken between 2003 and 2004. The value 100 billion I referred to was mentioned in the Cosmos Series in the early 80's. So there must be a source for that total number other than the Hubble's Ultra Deep Field.

My understanding is that at the time, observational astronomy was not yet equipped to measure the number of galaxies to any significant degree of accuracy, so it was largely a guess. One informed by observation, but that could have been much further off than it was.

I'm pretty sure the first solid evidence of a lower bound of around 100 billion galaxies really did come from Hubble (though with the Deep Field in 1992, not the later Ultra Deep Field, as I mistakenly said earlier).

 

[phyzguy]

If you ask, "How many galaxies are in the observable universe?" you need to be more precise about what you are asking. If you read the link that phinds provided, the reason the estimate went up is that they decided that there were many very faint galaxies in the early universe that have since merged into larger galaxies. So if you are asking, "How many galaxies can we see?" then the proposed increase makes sense. If instead you are asking, "How many galaxies are in the observable universe now?" you will probably come up with a smaller number. Of course, you then get into issues of how you define "now" in an expanding universe.

 

[Thuring]

Wilipedia:

In 2016, a study published in https://www.physicsforums.com/x-dictionary:r:'The_Astrophysical_Journal?lang=en&signature=com.apple.DictionaryApp.Wikipedia' and led by https://www.physicsforums.com/x-dictionary:r:'Christopher_Conselice?lang=en&signature=com.apple.DictionaryApp.Wikipedia' of the https://www.physicsforums.com/x-dictionary:r:'University_of_Nottingham?lang=en&signature=com.apple.DictionaryApp.Wikipedia' using 3D modeling of images collected over 20 years by the Hubble Space Telescope concluded that there are over 2 trillion (2×1012) galaxies in the observable universe.

 

[phinds]

Wilipedia:

In 2016, a study published in https://www.physicsforums.com/x-dictionary:r:'The_Astrophysical_Journal?lang=en&signature=com.apple.DictionaryApp.Wikipedia' and led by https://www.physicsforums.com/x-dictionary:r:'Christopher_Conselice?lang=en&signature=com.apple.DictionaryApp.Wikipedia' of the https://www.physicsforums.com/x-dictionary:r:'University_of_Nottingham?lang=en&signature=com.apple.DictionaryApp.Wikipedia' using 3D modeling of images collected over 20 years by the Hubble Space Telescope concluded that there are over 2 trillion (2×1012) galaxies in the observable universe.

And that statement in no way addresses the issues pointed out to you by phyzguy in post #22

 

[Thuring]

It provided the recent source for the 2 trillion estimate.

 

[phinds]

It provided the recent source for the 2 trillion estimate.

Right. Estimate of WHAT, is my point and that of @phyzguy