The Mystery of CMB Polar Anisotropy and its Implications on Cosmology

[profgemelli]

Hello, friends. I read that polar anisotropy of the CMB shows that the solar sistem is moving towards the Virgin constellation. This polar anisotropy is not something which is not going to cause some problems...

First question: Isn't this a sort of ABSOLUTE MOTION? i.e. we have found out that there exists a cosmological global rest reference frame, the one with respect to which the CMB has no polar anisotropy: it is a cosmological preferred frame, which all in all allows astronomers to say we are moving towards the Virgin. After all relativity has started loosing some appeal?

Second question: Is the zero polar anisotropy reference frame the classical ETHER? After all after Michelson and Morley other people confirmed that there is no light anisotropy, so there is no ether, or shouln't be... It seems to me that there is no light anisotropy but only in the CMB... this is a mistery to me.

Third question: polar anisotropy of CMB does it violates the cosmological principle or not? It seems to me that if the Universe should look the same from everywhere while different galaxies with diferent motions with respect to the zero CMB polar anisotropy reference frame have different polar anisotropies!

 

[marcus]

There is no contradiction. The ancient universe (say 300,000 or 400,000 years after start of expansion) was filled with hot gas.

As soon as you put a gas into the picture, this gives a rest frame. The frame in which the gas has no average overall motion, only random motion of particles.

When we detect the CMB we are looking at the glow of that hot gas. In effect we are looking at the ancient matter and the ancient light let's us define the rest frame of the ancient matter.

There is no contradiction of special relativity, or of general relativity. It is just as one would expect from adding matter (gas) to the picture.

So we naturally want to know the motion of the solar system relative to that.

I think the direction of motion is towards the constellation LEO, not Virgo, as you say.

 

[cepheid]

I think you mean "dipole anisotropy." In any case:

From Introduction to Electrodynamics 3rd Edition by David J. Griffiths:

Einstein proposed his two famous postulates:

1. The principle of relativity. The laws of physics apply in all inertial reference frames.

2. The universal speed of light. The speed of light in vacuum is the same for all inertial observers, regardless of the motion of the source.

The special theory of relativity derives from these two postulates. The first elevates Galileo's observation about classical mechanics to the status of a general law, applying to all of physics. It states that there is no absolute rest system. The second might be considered Einstein's response to the Michelson-Morley experiment. It means that there is no ether. (Some authors consider Einstein's second postulate redundant --- no more than a special case of the first. They maintain that the very existence of the ether would violate the principle of relativity in the sense that it would define a unique stationary reference frame. I think this is nonsense. The existence of air as a medium for sound does not invalidate the theory of relativity. Ether is no more an absolute rest system than the water in a goldfish bowl --- which is a special system, if you happen to be the goldfish, but scarcely "absolute.")⁵
____________________________________
⁵ I put it this way in an effort to dispel some misunderstanding as to what constitutes an absolute rest frame. In 1977, it became possible to measure the speed of the Earth through the 3 K background radiation left over from the "big bang." Does this mean we have found an absolute rest system, and relativity is out the window? Of course not.

 

[profgemelli]

I see this can be an answer to my first two questions; then let us assume that dipole anisotropy (my mother language is not English so I have mispelled it) is not a menace to special relativity, ok.

But is it a menace to traditional cosmology? Does its presence violate the cosmological principle? This principle says: "universe must look the same from everywhere", which is not, since any observer in the universe can measure his CMB and thus his local speed with respect to the cosmological fluid. Am I wrong?

 

[Calimero]

No, it is a pillar stone for modern cosmology. http://map.gsfc.nasa.gov/news/5yr_release.html" you can see some of the stuff that we have learned from it. It is truly amazing, really.

As for cosmological principle, if you subtract expansion, it is quite normal to expect that galaxies will have peculiar velocities. In fact those velocities are not so great, and they only influence the picture on smaller scale.

 

[bapowell]

But is it a menace to traditional cosmology? Does its presence violate the cosmological principle? This principle says: "universe must look the same from everywhere", which is not, since any observer in the universe can measure his CMB and thus his local speed with respect to the cosmological fluid. Am I wrong?

The universe is homogeneous and isotropic on large scales. Of course there are deviations from these symmetries on small scales -- just look around you. All observers move relative to the expansion (i.e. the CMB), since on small scales there are inhomogeneities. As we move relative to the CMB, we detect a dipole anisotropy. And, sure, another observer in a distant reach of the universe might measure a different dipole if they are moving sufficiently different than us. The cosmological principle simply states that on sufficiently large scales, these so-called "peculiar" motions of observers relative to the CMB should average out.

 

[profgemelli]

Thank you friends.

I still am doubtful about defining the dipole anisotropy something on a small scale: isn't it even bigger than the galaxy emission anisotropy, which in turn is bigger than the depured CMB anisotropies of the famous WMAP picture (the very tiny ones in which one sees the seeds of the galaxy clusters)?

 

[bapowell]

Why does it matter how "big" it is? Indeed, the primordial anisotropies are quite small in comparison.

 

[profgemelli]

You say: the universe still look the same from anywhere, but on large scales. But then you say that it's not important how big the anisotropy is. Then what defines small scale and large scale? In which sense the universe does still look the same from everywhere? I would say that if the component of the CMB which really looks the same from everivhere is very tiny while the "local motion" component is much larger, the CMB should violate the cosmological principle (forgive me my poor language).

 

[bapowell]

I did not say that it does not matter -- I asked *you* why it matters.

By "big" I am referring to the amplitude of the temperature anisotropy as a function of angular scale. The amplitudes of the anisotropies are Gaussian distributed (as far as we can tell.) The distribution of the initial density perturbations (which lead to the primordial temperature anisotropy) dictate the clustering of galaxies, and in principle, the peculiar velocities of all galaxies (although in an admittedly complicated, nonlinear fashion). So, the statistical distribution of peculiar velocities must be dependent on the statistical distribution of initial fluctuations. The large-scale homogeneity and isotropy of the universe is confirmed by measuring the mean temperature of the CMB, which happens to be much larger than the dipole. But there will be outliers. These outliers do not invalidate a highly-symmetric universe, but they must be statistically insignificant.

 

[profgemelli]

Perhaps I can understand, now: different dipole anisotropies in different places should compensate globally to give a mean null anysotropy. Am I right?
I guess this is a (legitimate) assumption, not something we can know for sure, is it?

 

[profgemelli]

Aaargh! I thought I got it a moment ago but now I am doubtful again! How could different proper motions give a null mean motion? No, this makes no sense, I must be wrong. It seems to me that dipole anisotropy of CMB means that it is a fact that the universe DOES NOT look the same from everiwhere. So this DOES violates the cosmological principle!

 

[bapowell]

I'm not sure why you are principally concerned with the dipole, since the primordial temperature fluctuations themselves destroy homogeneity and isotropy (afterall, they're called temperature anisotropies). According to the cosmological principle, the inhomogeneities should have the same statistical properties for all observers. One way of thinking about this is to imagine that each observer sees a universe chosen from a statistical ensemble of possible universes, and so each observer will measure a different universe. For example, while the distribution of galaxies is approximately uniform, there is a finite probability that an unusually large void exists somewhere in the universe...observers near this void certainly see a different universe (so much so that the cosmological principle might be very far from obvious). The same goes for peculiar velocities, and the size of each observer's dipole. These will indeed vary from observer to observer across the universe, but this variance must be consistent with the statistical distribution of the initial perturbations.

 

[profgemelli]

First of all, thank you for going on answering my humble (and gramatically rough) doubts and questions.

The reason why I am concerned with dipole anisotropy is that it seems to me to define a sort of absolute motion, and also to define a privileged reference frame in cosmological sense (i.e. those of null dipole anisotropy) whose picture of the universe seems to me very different from the others.

It is something which in my opinion is still not desiderable (if not critical) since somewhat contrasts the fact that motion has to be relative, not absolute, and the fact that the Universe should look the same from everywhere.

 

[marcus]

There is no contradiction. The ancient universe (say 300,000 or 400,000 years after start of expansion) was filled with hot gas.

As soon as you put a gas into the picture, this gives a rest frame. The frame in which the gas has no average overall motion, only random motion of particles.

When we detect the CMB we are looking at the glow of that hot gas. In effect we are looking at the ancient matter and the ancient light let's us define the rest frame of the ancient matter.

There is no contradiction of special relativity, or of general relativity. It is just as one would expect from adding matter (gas) to the picture.

So we naturally want to know the motion of the solar system relative to that.

I think the direction of motion is towards the constellation LEO, not Virgo, as you say.

... since somewhat contrasts the fact that motion has to be relative, not absolute,..

Our proper motion (the source of the doppler dipole) is relative. Just as it should be.
It is relative to the hot gas of the recombination era, and to the soup of photons released at that time, which still surround us (about 413 million photons per cubic meter)

That is matter. In Gen Rel the soup of light in which we sit can be considered to be matter. As soon as matter is put into the Gen Rel picture, there is motion (and rest) relative to that matter.

I said this already in post #2, and Brian Powell has been constantly saying the same thing. And you still seem not to understand. THERE IS NO CONTRADICTION.

Whenever you put some matter into the GR picture then it becomes meaningful to talk about being at rest relative to that matter.

Whenever there is some matter, there is always a preferred frame relative to it.

This preferred frame is in full accordance with Gen Rel. It is not forbidden to have a preferred frame if there is matter.

 

[profgemelli]

I have been thinking about it.

I understand that matter such as perfect fluid or something has a so-called comoving (or proper) reference frame... but CMB? Is it matter? Isn't it radiation?

Let me think: radiation has a Maxwell stress-energy tensor. This also has a comoving reference frame? We are talking about the "rest frame" of a fluid of photons?

Thank you in advance for your patience.

 

[bapowell]

Yes, a perfect fluid has a comoving frame. The CMB, ignoring its inhomogeneities, is well approximated by a perfect fluid. Reread Marcus's first post -- the rest frame of the CMB is the frame in which the fluid has "no average overall motion", i.e. it's the comoving reference frame of the fluid.

 

[gvgomez]

This discussion highlights an important issue. But the contradiction does not concern the
cosmological principle, as profgemelli still believes, but the anisotropy of light. The dipole anisotropy is interpreted in a natural way in terms of Doppler effect due to a source in motion relative to us, which is really a version of the Michelson experiment. It 'true that so far no one could ever deniedy the results of Michelson, but the CMB is an experiment as big as the universe and that has lasted billion years, it is obvious that it can show effects that our interferometers cannot see. I'm sorry for Einstein, but the ether, unfortunately, seems to exist.

 

[cepheid]

This discussion highlights an important issue. But the contradiction does not concern the
cosmological principle, as profgemelli still believes, but the anisotropy of light. The dipole anisotropy is interpreted in a natural way in terms of Doppler effect due to a source in motion relative to us, which is really a version of the Michelson experiment. It 'true that so far no one could ever deniedy the results of Michelson, but the CMB is an experiment as big as the universe and that has lasted billion years, it is obvious that it can show effects that our interferometers cannot see. I'm sorry for Einstein, but the ether, unfortunately, seems to exist.

You seem to be confusing two different effects. There is the Doppler effect (relative motion between source and observer changes the observed frequency of light), and then there is the effect that Michelson was hoping to see in his interferometer, which is that relative motion between source and observer would change the observed speed of light (which would in turn change the optical path difference between two arms of an interferometer). These are two different effects. What Einstein was saying was that the speed of light observed would be independent of relative motion between source and observer. He was not denying that Doppler shifts occur. The latter is a consequence of the wave nature of light of which he was well aware.

EDIT: Another way of putting it: the motion of Earth relative to the CMB rest frame is NOT a version of the Michelson experiment, unlike what you have claimed above.

 

[Imax]

We have found out that there exists a cosmological global rest reference frame?

It’s unlikely under special or general relativity.

Yes, a perfect fluid has a comoving frame. The CMB, ignoring its inhomogeneities, is well approximated by a perfect fluid. Reread Marcus's first post -- the rest frame of the CMB is the frame in which the fluid has "no average overall motion"

Agree :)

 

[profgemelli]

Thank you, to everybody. Your answers were useful and sometimes enlightening to me, and always were conforting to me (but that of gvgomez). I must confess I still have some residual suspicion about dipole anosptropy, which maybe I will try to better focus, but I now certainly feel more confortable with it. Thank you.