I was not certain where to post this but I figured here would be best. I came across this upon a web page:


Question - If light travels through space, why is space dark?
This is a question that has been around for centuries. It is attributed to
an astronomer Heinrich Olbers, and bears the name Olbers' (sometimes called
Olber's) paradox. Your inquiry motivated me to do a web search -- I thought
the answer was simple, I was missing something obvious, and a quick
refresher would supply a profoundly simple answer. I was badly mistaken!

The paradox arises by assuming 1. The universe is an infinite Euclidian
space. 2. The age of the universe is infinite. 3. Matter is uniformly
distributed. 4. The universe is static.

The terms "infinite" and "static" not literally, but sufficiently large to
be approximated by these terms. None of these assumptions is exactly true
and some "explanations" just say the assumptions are wrong therefore there
is no paradox. Sorry, but that is just avoiding the issue. It is clear that
we do not "see" all the radiation in the universe and if we could see
everything from cosmic rays to microwaves, the sky would be uniformly
bright, but that too is only a conjecture because at the present time we
cannot "see" all the radiation in the universe at all wavelengths
(energies). Ronald Koster has proposed a resolution which may be correct,
that says that we are shielded from radiation originating very far away and
gives a simple calculation that he contends resolves the paradox, but I'm
not an astrophysicist, so I am not able to assess the correctness of his
arguments.

I will be interested to see what other Newton BBS responders have to say
about this, too.
I do not think the answer is simple.

Vince Calder
================================================== =======

One day while lying in the cafe at my job after I had finished working it came to me, perhaps the answer is alot more simpler then we like to admit. So I thought why is it that in the warehouse there is light a plenty as well as in the cafe yet the cafe is so much brighter. Could it be cause the cafe is smaller there by less time for the light to travel...lol, ok that was a joke, but seriously could it not be that the reason it is dark in space and say light here on earth simply because of reflection? In space there isn't too much in the way of reflecting the light that is to illuminate an area. I sure if we surrounded the sun in whats that a dison sphere, (I saw that on Star Trex the next Generation) that we would illuminate the solarsystem quite nicely. The smaller that dison sphere would be the brighter it would be and or the more objects placed within that area the brighter it would be, hmmm?? Thoughts....

The one and only Eyajwhynsos

__________________
The limits limitless...."prove it?" is what is said

well take a half of a step of each step taken in life, and be living to you say when....

I think the simplest explanation of Olber's Paradox is in the framing of the question: Infinity.
If the universe is finite in size but infinite in age, then you get the paradox as stated. If the universe isn't inifinite in age, then this "problem" goes away.
The universe, as best we can tell currently, is about 13.7 billion years old, give or take 0.2 billion years.
My simple calculation, being that I'm not very good with math, shows that 13.7 billion +- 0.2 isn't close at all to infinity.

__________________

Feynman
>~~~~<
Science is a way of trying not to fool yourself. The first principle is that you must not fool yourself, and you are the easiest person to fool.

Religion is a culture of faith; science is a culture of doubt.

If the universe isn't infinite in age, how exactly does the "problem" go away? I mean now I gotta play my side for a bit, but if there is nothing to reflect light off of, then all you have in the case of outer space (the universe) are but points of lights. Even if these points of light(stars) are very close to each other(galaxies), you still would have and do have a very dark surrounding area. Take for instance a single light bulb hanging in a large room, the larger the room is the less lit it would appear, the smaller the room is the more lit it would appear, yet if there were no walls or any type of surrounding like you have out there in space what would the light reflect off of? Well I really am only theorizing, does anyone have an alternative to either one of these post?

The one and only Eyajwhynsos

__________________
The limits limitless....&quot;prove it?&quot; is what is said

well take a half of a step of each step taken in life, and be living to you say when....

I think the theory goes something like this: If the universe is infinite in size with an infinite number of stars, every point in the night sky you look at will have a star or galaxy in it, although some will be closer and others will be much further away.

But if it’s not infinite, you can look at a lot of points in the night sky and see nothing at all.

Because space is empty.

__________________
Join the Illuminati

I read somewhere that it has something to do with expansion...

Unfortunately, I forgot both the details of why that is, and the name of the book where I read that. -_-

Are you talking about the fact that the space that we can see is filled with a lot of stars, and they are emitting light in all directions, so why is the sky black?

We can’t see light rays from the side. They can pass right past you, in front of you, and you can’t see them. They’ve got to enter your eye directly in a straight line.

It like you can’t see laser light from the side. If you blow smoke at laser light, you will see the beam appear to be “lit up” but that’s because part of the laser light hits the smoke particles and some of that light is reflected by the smoke particles directly into your eyes. You can’t see light from the side. So all those light beams zooming around in space can not be seen unless they travel directly into your eyes. This leaves the background of the stars totally black.

http://en.wikipedia.org/wiki/Olbers%...ed_explanation

Not sure how accurate it is, but I'm sure someone could correct or clarify some points.

Just in case anyone is curious about Dyson spheres.

http://users.rcn.com/jasp.javanet/dyson/

Well of course the other things is that we've found that areas of space that appear "empty" aren't when you have powerful enough optics. Often the number of photons of light are just so spread out that only a few reach us and so we need super optics to see anything there. Take the Hubble deep field for example. Red shifting of light into the infra-red spectrum could also account for some of it. too.

__________________
Howling from the Shadows

It must be fun to lead a life completely unburdened by reality. --- JayUtah

You can't reason an irrational person out of an irrational belief. --- Noclevername

Apollo: The History and the Hoax
Enter the World of Athran

Halcyon Dayz says: Because space is empty.


My sentiments exactly, not much to reflect off of.

__________________
The limits limitless....&quot;prove it?&quot; is what is said

well take a half of a step of each step taken in life, and be living to you say when....

a lot of material shows up in the infrared...(at least within our galaxy)?

***********************************************
image centred on draco

draco

240 micron (infrared) sfd
100 micron (infrared) iras
60 micron (infrared) iras

60 * 45 degrees fov
galactic/gnomonic projection
***********************************************
image centred on m45

m45

240 micron (infrared) sfd
100 micron (infrared) iras
60 micron (infrared) iras

20 * 15 dg fov
galactic/gnomonic projection

**********************************************
image centred on cygnus-a (invisible in this shot)

cygnus-a

240 micron (infrared) sfd
100 micron (infrared) iras
60 micron (infrared) iras

30 * 22.5 dg fov
galactic/gnomonic projection
**********************************************
image centred on iris nebula

iris nebula

red = 100 micron (infrared) iras
green = 60 micron (infrared) iras
blue = 25 micron (infrared) iras

30 * 22.5 dg fov
galactic/gnomonic projection
**********************************************
below lockman's hole

red = 100 micron (infrared dust) iras
green = ?nanometre (ultra violet) ?
blue = 3/4 kev (xray) rosat

60? * 45? dg fov
galactic/gnomonic projection
**********************************************

images best viewed as desktop wallpaper.

This seems to be correct, and it seems possible to me that the universe could be “infinite”, but there could be a viewing-distance limit where the incoming photons from a distant galaxy are so spread apart by the time they reach us, we can’t see them. This might allow for a dark sky in an infinite universe filled with an infinite number of galaxies, so this phenomenon could overrule Obler’s Paradox.

The universe, in my opinion is most probably infinite, but it's not infinitely old (not an opinion, but based on evidence).
You don't need a viewing limit where the incoming photons "spread out", the viewing limit is that they are farther away in light years then the universe is old, and hasn't reached us yet.

__________________
"Ignorance has caused more calamity than malignity" H.G. Wells

"Getting lost is part of exploring." Uniqua in "Backyardigans-Heart of the Jungle"

"Trying to wrap my head around creationist astronomy is like trying to ride a unicycle around a Moebius strip: it’s off-balance, physically impossible, full of one-sided arguments, and in the end you don’t go anywhere." Phil Plait

the universe is dark but isn't dark;

of course if we had the capacity, as Humans, to see ALL wave-lengths of the electromagnetic spectrum at the same time, we'd be blinded!!

or could we adapt over thousands of millennium to see it all, as a whole, the Universe i mean, its electromagnetic spectrum. hmmm......

Olber’s paradox assumes a more or less uniform distribution of stars in space. Think of the stars we see as distributed around us in nested thin spherical shells. Although the light flux reaching us from any star in a shell is inversely proportional to the square of the radius of the shell, the number of stars in a shell is directly proportional to the square of the shell radius. Therefore the total light flux from a shell is independent of the shell radius. With a sufficient number of shells the total flux reaching us should produce a night sky as bright as day.

It seems that the fundamental error in Olber’s model is the assumption that stars are uniformly distributed in space. Stars are not uniformly distributed in galaxies. Stars are even less uniformly distributed in galaxy clusters and still less in superclusters. The larger the cosmological object we observe the smaller is its spatial density of stars. The spatial density of stars in a galaxy cluster is orders of magnitude smaller than the spatial density of stars in a galaxy. The voids between galaxies in a galaxy cluster are much larger than the galaxies. Every step up the cosmological structure ladder introduces voids much larger than the objects being separated. The cosmological structure ladder assures a non-uniform distribution of stars in space.

Will this resolution of Olber’s paradox draw fire from Cosmological Principle of Homogeneity supporters?

Why?

I mean, by how much does the total, integrated energy of all EM, received here on Earth, exceed that which our eyes are sensitive to?

First, spatial inhomogeneity, in a static, infinite universe, doesn't necessarily resolve Olbers' paradox - pick a line of sight, any line of sight. Now, for a static, infinite universe, with spatial inhomogeneity, outline how such a line of sight cannot, ever, intersect with the surface of a star (intersecting with a cloud of gas or dust doesn't help; such thing will be, in a static, infinite universe, in thermal equilibrium with the incoming EM ... to them).

Second, the universe's observed spatial inhomogeneity declines as the scales get bigger. IOW, as you take larger and larger 'chunks' of the universe, it gets more and more homogeneous.

Nereid wrote:The issue in Olber’s paradox is not the mere fact that every line of sight in an infinite universe would end on a star. Olber’s paradox requires that the spatial density of stars should not decrease with distance from the observer, otherwise the inverse square law might overwhelm the light flux per shell and result in a finite, small summed series of flux contributions from an infinite number of shells. Such a sum could not produce Olber’s nighttime sky as bright as the sun.

Heinrich Olber wrote about the paradox in 1826. At that time the local galaxy was thought to be the universe. Every improvement in telescopes revealed more stars in previously dark spaces. The spatial distribution of stars in the local galaxy was not known. Hence the assumption that on a large scale stars might be uniformly distributed in space. After Hubble proved that the Andromeda galaxy was a great collection of stars very distant from the local galaxy, the universe was thought to be a great sea of galaxies. It was then thought that on a large enough scale galaxies would be uniformly distributed in space. Later it was discovered that galaxies clumped together as clusters and that clusters clumped together as superclusters.

Whether the inverse square law wins or not depends on the spatial distribution of stars as a function of distance from the observer. We are about 2/3 out on one of the spiral arms of our galaxy. The nearest star in the night sky is Alpha Centauri at a distance of 4-1/3 lightyears (LY). All the bright stars of the night sky are in our neighborhood of the local galaxy. Since our galaxy is about 100,000 LY in diameter the inverse square law reduces most of its light to the diffuse, dim light of the Milky Way. The light contributions of the few dwarf galaxy satellites of the local galaxy can be neglected.

The nearest member of the local group of galaxies is the Magellanic Cloud (MC) at a distance of 179,000 LY. The local group has a diameter of about 10,000,000 LY. The Andromeda galaxy (M31), a large neighbor in the local group, is at a distance of 2,300,000 LY. M31 appears to the naked eye as a patch of very dim light. (How about the recent stunning Spitzer picture of M31 and its dust!)_ Other neighbors are at comparable or greater distances. The farthest member of the group is Sextans A, at a distance of about 4,730,999 LY.

Let us examine the implications of the data in the preceding two paragraphs. The densely star populated region of space nearest to us is the local galaxy. The light it contributes ranges from bright for close stars to very faint over a distance of 100,000 LY. The next dense collection of stars, MC, is 179,000 LY away. Its light, using the inverse square law, is about 8/100 as intense as the faint Milky Way. Keep in mind that the local galaxy and MC are not shells of stars surrounding us; they occupy only a small portion of the sphere of space out to their distance. The light from Sextans A, at the far side of the local group, is about 1 /10,000 as intense as the faint Milky Way. The local group, consisting of some 30 galaxies, also is not a shell of stars surrounding us. The first 4,730,999 LY radius volume of space shows an enormous decrease in star density of space with distance. If we take the light of the local galaxy as our unit candle (1 candle), then it can be readily shown that light from the remaining members of the local group contributes only 0.135 candle.

In order to produce a night sky as bright as the sun, the spatial density of stars would have to start to increase enormously further out to compensate for the density decreases so far, and for the next inverse square law reductions. At what distance might that occur? It will not occur at the next step up the structure ladder. That next step, the local supercluster, introduces voids larger than the clusters. Star density drops again, and that brings us to distances where the inverse square law really gets important.. As for the other superclusters, enormous voids are between them. Distances get into the billion LY range with inverse square law reductions that really take a toll on the light we receive from out there. It is not likely that superclusters are ultimate structures, or building blocks of the universe. As Bruce medalist astronomer C. V. L. Charlier first suggested, matter is clumped in space at all scales. Clumping introduces voids.

Nereid also wrote: Even if such increasing homogeneity were true (which it may not be), it is irrelevant to the Olber’s problem. To say that the further out you look the more homogeneous it seems does not mean that most of the light received on earth comes from those distant homogeneous seeming regions. The light from the nearest regions, which is least diminished by the inverse square law, comes from a very heterogeneous distribution of stars or galaxies. Even if the most distant regions seem more homogeneous, that does not preclude the possibility that densities might decrease beyond those regions.

In The Density field of the local Universe by Will Saunders, Carlos Frenk, Michael Rowan-Robinson, George Efstathiou, Andy Lawrence, Nick Kaiser, Richard Ellis, John Crawford, Xiao-Yang Xia, & Ian Parry, Nature, vol 349, 3 January 1991, page 32 data from the Infrared Astronomical Satellite were used to map a part of the Universe on a whole sky basis. In addition to the previously well known 19 superclusters and 11 voids, 8 new superclusters and 4 new voids were identified. The following parenthesized phrase appears in the abstract of the article: “(the voids are larger than the superclusters but depart less from the mean density).”

In #16 I wrote That is true for an atom, for the solar system, for a galaxy, for a galaxy cluster and for a supercluster. That introduction of voids means that density drops dramatically at every step up the ladder. The local galaxy is much less dense than the star systems comprising it. A galaxy cluster is much less dense than the galaxies comprising it. A supercluster is much less dense than the clusters comprising it. An ultracluster, if there were such a thing, would be much less dense than the superclusters comprising it. As you take larger and larger ‘chunks’ of the universe the spatial density of stars decreases ever more dramatically.

The ‘Density field …’ article gives the mass densities of the identified superclusters and voids. The 27 supercluster densities range from 0.13 to 5.49. The 15 void densities range from 0.00 to 0.41. Even at that scale things look far from homogeneous.

Another problem for the Cosmological Principle of Homogeneity is that with every step up the cosmological ladder the membership is less than that at the preceding step. There are fewer stars than atoms, fewer clusters than galaxies, fewer superclusters than clusters, and so on. Considering the sample size, how can we know that the range of superclusters presently identified is representative of those that are still unknown? And how about the density at the next step up the ladder, or is there some reason that either superclusters or the next higher structures are ultimate structures, i.e. ‘building blocks of the universe’?

It seems to me that the inverse square law doesn't matter in this case. As long as the line of sight at any angle falls upon the surface of a sun, the sum of the light from each of them would be equal to the average brightness of a sun at the point of observation, regardless of distance. Now if half of the lines of sight fell on dead stars or black holes, then the average brightness would only be half as great. In other words, the distribution of stars does not matter, even with the largest voids imaginable in between.

Gray, the possible exception to that idea is if the scaling goes on ad infinitum. In that case (and I think this is what Richard is implying) the density wll approach zero, meaning it will be zero. So there would be no object within any given line of sight. This has come up before in threads on fractal cosmology and on Olber's paradox.

__________________
As above, so below

Grav wrote Grav, the only stars whose light is visible to the naked eye are in the local galaxy and four nearby galaxies. The light from the remaining 26 galaxies of the local cluster is completely invisible to the naked eye. Although the local galaxy is far brighter that those four nearby galaxies, let’s assume that they are equally bright and assign each of them a brightness value of 1.00. To the invisible galaxies we assign a brightness value of 0.00.

The average brightness value is 4/30 or 0.14, less that 1/7 the brightness of the local galaxy, most of which is really pretty dim light. If we were to include the light of galaxies of the all other observed clusters, all invisible to the naked eye, the average brightness would be very close to zero. That would mean that the sky is dark at night.

Jens], [nit pick] for an infinite universe with an endless scale of structures the density would indeed approach zero, but it never could be zero. Infinitesimally small is not the same as zero. Only a universe with zero mass could have zero density. [/nit pick]

OK, I stand corrected. But it wouldn't matter, would it? If the limit approaches zero, that would still be good enough to get rid of the issue of Olber's paradox, wouldn't it?

__________________
As above, so below

Jens wrote: It might not be good enough. That the terms of a series approach zero does not ensure convergence of the sum.

Consider the series 1 + 1/2 + 1/3 + 1/4 +...

Terms of that series approach a limit of zero. The sum, nonetheless, is divergent and approaches infinity. Resolution of Olber's paradox requires a sum that converges.

I'll come back to this later.The information in these last few paragraphs have been made obsolete, by 2dF and SDSS.

As larger and larger volumes of space have been probed, the average density has indeed remained pretty much the same, and the degree of inhomogenity declined. IOW, there are no 'super-voids', or 'super-superclusters', in the sense that these previous paras imply. Indeed, Richard J. Hanak's summary is somewhat misleading, in the sense that the way in which luminous matter clumps isn't in the neat hierarchies described. For example, there's more mass in small clusters and galaxy groups than there is in rich clusters, and superclusters sometimes comprise looser clumps than clusters do.

In any case, all lines of sight end on the CMB (unless they end on a luminous star or nebula, or a dark cloud of dust or gas). Further, no sight-line is longer than ~13 bn light-years (care needed in interpreting this!), and all objects beyond the Local Group (and some galaxies in the Virgo cluster) are redshifted.

Further, no sight-line is longer than ~13 bn light-years

Well to be fair, we don't actually have a telescope that can see much further, though there are currently several are in the making that might be able too. We are seeing a little over 13 bn years and there are galaxies already well formed, so what we need is to be able to push beyond that point a see when they were forming at the least, but for that we're going to have to wait on the VLT and others to get fully operational.

__________________
Howling from the Shadows

It must be fun to lead a life completely unburdened by reality. --- JayUtah

You can't reason an irrational person out of an irrational belief. --- Noclevername

Apollo: The History and the Hoax
Enter the World of Athran

Phantom Wolf, thanks for being fair.

Nereid wrote: In those paragraphs of #19 I referred to the article by Saunders et al. Have the number of superclusters and the number of voids changed? Have the densities of the superclusters and voids changed?

Here is a graph of galaxies out to 2 billion LY from 2dFGRS.
http://mcp1.anu.edu.au/~TDFgg/Public...iew/sld067.htm Does white space represent empty space in the graph? Why is there more of it beyond 1.5 BnLY than before?

Nereid also wrote: If superclusters and their voids are still in the picture and there is to be no next step, are clusters or superclusters the ‘building blocks of the universe?’ Or does the universe become amorphous beyond the superclusters?

As for hierarchies, here is what Wikipedia has to say in “Large Scale Structure of the Cosmos”:If “there's more mass in small clusters and galaxy groups than there is in rich clusters, and superclusters sometimes comprise looser clumps than clusters do”, where does the homogeneity begin to appear?

My approach to resolution of the Olber’s paradox depends on the general validity of the stepwise increase of empty space between the cosmic structure levels, and the consequent stepwise decrease in density of stars. Is it wrong to claim that the spatial density of stars is less in a galaxy cluster than in the galaxies that comprise it, or less in a supercluster than in the galaxy clusters that comprise it, or less in all superclusters and all voids than in all superclusters alone?

Nereid also wrote:Redshift may not diminish naked eye visible light as much as is thought. After all, shorter wavelength radiation (UV, etc.) could be made eye-visible by redshift. That might compensate for light redshifted to IR or longer wavelengths. However, redshift is not needed to resolve the Olber’s paradox. There is a fundamental flaw in the paradox model.

Let us assume that the uniform density of stars assumption in Olber’s paradox is its fatal flaw. Therefore, let us be willing to conservatively discount any other reasons that might account for a dark night sky, such as redshift, tired light, or age or finiteness of the universe. Let us conservatively assume that all stars are equally bright in naked eye wavelengths, and even that all galaxies are equally bright. Let us also abide by the infinite number of shells idea of Olber’s assumptions, which implies an infinite universe.

The original Olber’s paradox assumed an infinite number of concentric, star-containing shells surrounding the earth. All shells have the same constant thickness, and the spatial density of stars is taken as everywhere the same. The light emitted from any shell is proportional to the number of stars in the shell. The number of stars in a shell is the product of its volume and the spatial density of stars. The volume of the shell is proportional to the square of the radius of the shell. The emitted light, then, is proportional to the square of the shell radius.

The light received from a shell is proportional to the emitted light divided by the square of the shell radius. Thus, the light received from any shell is independent of the shell radius and is the same from all shells. The sum of any series of constant terms increases without limit as the number of terms increase without limit and we should have a night sky brighter than day. Note that the spatial density of stars was a part of those constant terms.

However, Olber’s paradox does not even require a constant spatial density of stars in space to produce a light filled night sky. Consider the sum of light contributions of Olber’s shells if the density of stars in the second shell were half that of the first shell, the density in the third shell were one third that of the first shell, and the density in the nth shell were 1/n the density of the first shell. The only change we have made to the original version of the paradox is to the star density distribution. Now the received light from a shell is inversely proportional to the number of the shell. The sum of light received from those shells is the product of a constant and the sum of the series 1 + 1 / 2 + 1 / 3 + 1 / 4 + …. + 1 / n. That series, the harmonic series, is divergent. It’s sum increases without limit as n increases without limit. Again we should have a night sky brighter than day.

To have a dark night sky, all that is required is that the star densities of the shells form a convergent series. The analysis of contributions from the local group of galaxies presented in #19 indicated a strong convergence. Of some 30 galaxies in the group, the nearest galaxy contributes not 1 / 2 the intensity of the local galaxy, but 8 / 100 the
intensity. The thirtieth, and farthest member of the group contributes not 1 / 30, but 1 / 10,000 the intensity of the local galaxy.

The farthest member of the local group, Sextans A is at a distance of 4,730,999 LY. The next nearest cluster, the Virgo cluster is at a distance of 52,000,000 LY, more than ten times the distance of Sextans A. The contribution from a galaxy in the Virgo cluster at that distance would be about 8 millionths that of the local galaxy. Conservatively, let’s assume that all of the Virgo cluster’s galaxies are at that distance. The Virgo cluster contains somewhere between 1300 to 2000 galaxies. Let’s take the higher number 2000 and with extreme conservatism take them all to be at the same distance (53,000,000 LY). We could then expect the entire Virgo galaxy to contribute 16 one-thousandths the intensity of the local galaxy. Add that to what we previously estimated in [b]#19{/b] for the local cluster, and the brightness of the night sky is only 1.151 times as bright as the local galaxy. The sum of the series, despite extreme conservatism, still seems to be strongly convergent.

We need not carry these calculations to the next level. The distances indicate what we can expect: Magellanic Cloud 179,000 LY, Virgo Cluster 52,000,000 LY, Hydra-Centauri Supercluster 150,000,000 LY, Horologium Supercluster 905,000,000 LY.


There is another conservative feature of the preceding analysis. Not all of the light from the various structures is received on the night side of the earth. Some of it is received on the day side of the earth. If the night side of the earth faces away from the local galaxy, most of that great potential source of light is invisible. Not all of the local group or all of the local supercluster is observable on the night side of the earth.

The most serious conflict between Olber’s paradox and reality is the divergent series of contributions from the shells implicit in its non-realistic assumption regarding the spatial distribution of the density of stars. That is the resolution of the Olber’s paradox.

If the star density of the universe were constant, and we were to observe the luminosity of the stars at some distance, the luminosity of each would decrease with the square of that distance, but the number of stars on the spherical suface (with a defined thickness) for a distance is also proportional to the square of the distance. So multiplying the luminosity times the number of stars at any distance (for a given thichness) is a constant. This would imply infinite total luminosity at any particular point, but the line of sight can only reach as far as the first sun it reaches. The rest are "hidden" behind it. As a result, the night sky should be a bright as the surface of an average star. If the average star was only 1/100 as bright as ours is on the surface, then the night sky should be 1/100 as bright as the surface of our sun. If black holes and other dark matter blocks some of the lines of sight, then this would also be a factor.

Now if the star density of the universe is not constant, which it certainly isn't, since there are galaxies of stars and then clusters of galaxies, we would indeed have a convergent or divergent series, but only until we reach the void. At that point it begins again. In other words, if the star density is about constant, but only for galaxies, and the galaxy density is constant, but only in galaxy clusters, and so forth, as with fractals, then this would still all add up to the same thing. The only way we could end it completely is if we were at the center of the greatest density, which decreases hereafter never to rise again, which is unlikely, or to create an endless void. But since we can see these patterns repeating to billions of light-years away, even if the universe were a void from there on, it wouldn't be enough to dim the brightness by more than a mere fraction.

grav wrote: Don’t switch horses in midstream, grav. Olber’s paradox deals only with the density of stars in space. In his day they didn’t know about galaxies, let alone galaxy clusters. After all, even in galaxies it is the stars that emit the light.

The void is a relatively deep region of star free space. When we reach the void, the next stars are much farther away then the last. That added distance from earth greatly diminishes the light of those next stars. After the void it is not the same thing that begins again. It’s a continuation of what has gone before.

grav also wrote: You have found the way to end it, but it is not unlikely. It is exactly what we find. The shell closest to us has the least volume of all shells and contains the local galaxy. The density of stars in that shell is greater than in any other shell. That shell is next encased in a series of shells containing no stars. The next shell that contains stars is the one that contains MC. Then there are more shells containing no stars, and so on out to Sextans A. After that we are off to the next cluster. And after that, as before, the density of stars in a shell never rises again.

Here is a well known convergent series, the geometric series:

1 + 1 / 2 + 1 / 4 + 1 / 8 + …+ 1 / (2 ^ n)

As n increases without limit the sum of the series approaches 2 as a limit.

The sum of the very conservatively calculated light contributions in #19 and #27 reached a value of 1.151, and that includes the huge Virgo cluster. From the look of things, the remaining terms of that series will be hard pressed to raise the total light contributions to 1.200.

In the geometric series, the sum of all terms excluding the first is 1.000. In the series of light from the shells, the sum of all terms after the first will be less that 0.200. That series converges somewhat faster than the geometric series. If some of the conservatism in the calculations is removed that 0.200 might become 0.020 or 0.002. But even 0.200 means that the night sky is dark.

It's not so much that our understanding of the local universe has changed (though there is certainly a lot more data, and a lot more accurate data), rather that, with much deeper and much wider, systematic surveys in hand, far more detail about the large-scale structure of the universe has become possible.

A good example is this paper, based on 2MASS (my thanks to turbo-1 for bringing this to BAUT members' attention).

In particular, notice how the 3D distribution of galaxies and clusters is much less clear-cut - there are sheets ('great walls'), voids, filaments; 'superclusters' come in a wide range of shapes and degrees of concentration; and so on.

This lead to a change in the way large-scale structure is described - less superclusters, voids, and so on; more P(k).Kinda ... it represents a lack of galaxies with measured redshifts and positions. There are almost certainly galaxies in 'the white spaces', but they are too faint to have been selected for measurement in 2dF, or are 'behind' a foreground object (fewer of these), or a spectrum was taken, but it was too noisy to measure a redshift, or ...

The increasing amount of 'white space' is almost certainly primarily a selection effect - the further away a galaxy is, the brighter (intrinsically) is has to be to be selected in 2dF (and so there are fewer and fewer such galaxies, as z increases).Well, that's kinda the point. There is no discontinuity, no point of inflexion in a curve, {whatever} that you can point to and say "homogeneity sets in here".

As you take larger and larger chunks of the universe, and compare them with the same sized chunks from nearby (or not), the variation between chunks gets smaller and smaller. That's what P(k) shows - a smooth curve, with the degree of homogenity increasing as the scale increases.Yes, in the sense that each unit (galaxy, cluster, supercluster) occupies only a fraction of the space in its relevant scale.

So, it's not so much "what's the density of stars in galaxies?", rather "what's the density of stars in (random) chunks of space 50k pc in radius?"

Similarly, not "what's the density of galaxies in clusters?", rather "what's the density of galaxies in (random) chunks of space 1 Mpc in radius?"

Once you get above, say, 1 Mpc*, in scale, the average (space) density of stars becomes pretty constant (in an OOM sense).It's actually a lot more complicated than this.

There are certainly stars which emit far more energy in the UV (i.e. blueward of the optical) than in other bands - OB, Wolf-Rayet, white dwarfs, ... So when we see such redshifted, they will be 'brighter'.

However, such stars are almost entirely absent in elliptical galaxies (in the sense that they make up but a tiny proportion of such galaxies' total EM output), and the universe if filled with ellipticals! How does it work out, in fact?

Well, so far the record redshift for a galaxy is ~6, and such galaxies are invisible in the visible (they are 'seen' only in the NIR waveband), so the extent to which galaxies bright in X-rays (in their rest-frames) are also bright when seen at sufficiently high z, is academic (today).

But the reason for the optical dimness of high z objects has to do with the Lyman limit, the (EUV) region where neutral H absorbs any photons that happen to be wandering around.And I think you can see that this approach fails in several respects:

-> the space density of stars doesn't keep falling, beyond about 1 Mpc*, so each 1 Mpc shell added contains stars in proportion to the shell volume (more or less)

-> the highest redshift galaxy (= the furthest away) so far observed has z ~6; maybe we will one day 'see' galaxies, or stars, with a z of ~10 or 20, but not much further (so shells from there to infinity contribute nothing, from starlight)

-> (all?) galaxies with a z >~6 are 'Lyman dropout' galaxies - we detect no light from their (restframe) EUV

-> in the concordance cosmological models, all (photon) sightlines end on the surface of last scattering, which we see as the CMB (which has a z of ~1100).

So, the real (observable) universe is finite (shells beyond ~13 bn ly add nothing to the light we see), and redshifting does reduce the amount of light we see, from distant objects (cf what you'd expect from a flat SED and inverse square).

*Don't quote me on the actual number; I'd have to do some checking to see if it's more like 300 kpc or 3 Mpc or 30 Mpc, but the principle is right.