Just wondering: there is (I believe) a reference frame in which the CMB appears the same in all directions. In all other frames the CMB is Doppler-shifted to the red and blue in opposite directions. This “privaliged” reference frame is somewhat disquieting to this fan of Relativity. Anyone care to try to explain how this works and what its implications are?
-b
ps- sorry about the spelling. when is this place going to get a spell checker?
In an empty universe there is no preferred reference frame, but as soon as your universe has something in it, you can refer your motion to the motion of the other mass(es). The privileged reference frame is the frame in which the average motion of all the mass in our region of the universe is zero ( a sort of universal center-of-mass frame).
It’s called your brain.
Where’d you hear that the CMB is doppler shifted for most reference frames?
That would seem to majorly go against principles of relativity.
There shouldn’t be anything akin to a “center-of-mass” frame for the universe, as the universe does not have a center.
Eeenteresting. Reading about that now-- must be a reason why it isn’t contrary to relativity!
IANAP, but I’ll give this a shot. A frame in which the microwave background appears the same in all directions is known as a “comoving” frame. To every point in space there corresponds a unique comoving frame. However, this does not pose the problem in general relativity that it would if something similar were the case in special relativity.
In general relativity, inertial reference frames are generally only locally equivalent. No strictly local experiment can distinguish one frame from another, where local means that the experiment must take place over a sufficiently small region of space that curvature is negligible.
When you are looking at the light from distant galaxies, you are essentially performing an “experiment” on a region at least many thousands of light-years across, across which curvature (or at least expansion) is substantial. It should not be surprising, then, that one frame can distinguish itself as in some sense preferred to all others.
Probably one of the real physicists on this board can explain this better, but this might help.
But there is more to this than that. According to this, we are not the reference frame for which there is no doppler effect. It would seem that we can determine the direction of expansion.
From the little of this that I have just read, it sounds like 2 locations in the universe sufficiently separated and at the right angle from each other could comminicate with each other and triangulate where the center of the expansion is.
Now that seems to go against what I was taught. Perhaps I am incorrect in my interpretation, but I would definitely be curious about that.
I dunno if you are interpreting the COBE experiment correctly. I believe the angular variation in the frequency of the cosmic microwave background radiation is attributed to nonuniformity in the universe after the Big Bang at the time of radiation-matter decouping i.e. lumpiness in the fireball. What makes you interpret some or all of the dipole component of this angular variation as a Doppler shift?
Well, it is being stated that way in a number of references that I am finding. I can’t vouch for them, but that is where the interpretation is coming from.
Yep, the dipole moment (i.e. the fact that CMB is red-shifted when you look in one direction and blue-shifted in the other) tells us how we’re moving wrt the comoving frame. If you subtract out the dipole moment, the left-over variation tells about the lumpiness in the early universe (and so tells about the seeds of galaxy formation). But this says nothing about the “direction of expansion”. The expansion is occurring in all directions at once - there is no “center” of the expansion.
Check out this thread:
http://boards.straightdope.com/sdmb/showthread.php?threadid=109913&highlight=expansion+universe
No need to be an ass about it. I was feeling too lazy yesterday to cut and paste my post into Word to check the spelling. Sue me.
I do not think this is correct. If the universe is unbounded (I believe that is the right term–has no edges, anyway), there is no “center” to the expansion. Everything is expanding away from everything else with a speed proportional to distance.
I did a report on this a couple of years ago! Here’s what I remember.
The CMB is isotropic to about one part in 1000. At that level, you have anisotropies due to the fact that we’re moving with respect to the CMB reference frame. About one part in 1000 mean’s we’re moving at about 1/1000 c with respect to it. I think the actual number is between 300 and 600 kps.
You can account for this anisotropy. If you do, you again see an isotropic CMB, down to the level of about 1 in 100,000. These tiny-level fluctuations are what’s due to the distribution of the Universe at the last scattering surface (Big Bang + 300,000 years). There is also at this level an equatorial anisotropy caused by the galaxy, but that’s uniteresting, and you can remove out, so people do.
Now, why does this not violate Relativity? Easy. All Relativity says (I believe) is that Maxwell’s Equations hold in all reference frames. Does this contradict that? No. You could also say that we’re moving with respect to the Andromeda Galaxy. Just because the CMB is cosmic in origin doesn’t make it special in this respect.
IANACosmologist, but simply a lowly planetary astronomer, but I’ll take a whack at this. (I’m repeating what some other posters have said. )
The CMB is the heavily redshifted spectrum of the early, hot universe. The CMB is redshifted by the expansion of the Universe. (This is not technically a Doppler shift, but that’s pretty much a side issue for this discussion.) If the observer is just riding along with the expansion of the Universe, in the Hubble flow, as we say, then there will be no dipole moment for that observer, just the uniform redshift caused by the expansion of the Universe (which, as I said, isn’t really a Doppler shift).
Keep in mind that the expansion of the Universe is an expansion of space, not an expansion through space, so there is no point where the Hubble flow is zero–or, conversely, any point that is comoving with the Hubble flow has a velocity of zero with respect to the CMBR, so you could define any point as having a Hubble flow of zero, and them measure the Hubble flow of all points realtive to your chose reference point.
None of these points are “special” in the sense of violating relativity, because you can still do any physics (even physics related to the CMB) in any inertial frame. Physics done in a frame that is stationary relative to the Sun is no more or less valid than that done in a frame moving by at a constant velocity of .8494736c, and the CMB isn’t any different.
We observe a dipole moment(s) in the CMB because we orbit the Sun, the Sun orbits the center of the Galaxy, and the Galaxy is perturbed from the Hubble flow by the gravitational influence of other galaxies. The biggest effect is that the Local Group (containing the Milky Way, Andromeda, and some smaller assorted galaxies) is moving at 600 km/s relative to the CMB, toward the mysterious Great Attractor.
Okay, even tho I’m not the OP-er thanks for restoring some focus for me. I used to do relativity banter here a lot, but I haven’t been playing on this board for awhile. 
So here’s what I came away with, which as makes sense in my head now.
Across points in space, any “comoving” frame will see no CMB doppler shift. Two comoving frames at distance from each other will only see a Hubble red shift between each other. To use the old fashioned balloon analogy (yes, I know that the closed universe model has fallen out of favor), a comoving frame sits on the balloon, and a CMB doppler shift occurs only when the ant is moving around on the surface (local interactions, primarily gravitational).
Is that what everyone agrees?
[hijack]
This is how I have always understood it. However, I have encountered considerable confusion on this point. Do you mean by this that all space is expanding uniformly? Do you know of a cite that goes into this point in more detail?
Thanks!
[/hijack]
More or less. There are some local perturbations due to gravity, but if you look on a large enough scale, all the matter appears uniformly distributed, and space is expanding uniformly everywhere.
It’s always possible to define something as moving relative to some object, by the way, and some objects (particularly large ones) make more convenient reference frames than others. The hot plasma which we are now seeing as the cosmic microwave background is a very large object, but it’s still just an object, and the laws of physics don’t particularly care any more about it than they do about anything else.
And by the way, Achernar, the contribution from the Galaxy is actually the largest effect (several orders of magnitude greater than the dipole anisotropy), but it is subtracted out.
Just to make sure I understand, is it fair to say, putting any gravitational effects to the side for the moment, that a 1 cubic meter volume of empty space will, over time, become a 2 cubic meter volume of empty space?
Okay, I believe you. Maybe then you could explain this image of the CMB to me, which is where I got the idea that the dipole is subtracted before the disk interference. According to the page that that image comes from, “The top map includes the dipole (due to the motion of the Earth, Solar System, and Galaxy through the universe) and Galaxy; the dipole has been subtracted from the middle map; and both the dipole and Galaxy have been subtracted from the bottom map.”
Yup. Trippy, ain’t it?
Man without evidence should never argue against man with evidence, so I’ll bow out on this one until I can check my sources. Even if the dipole is larger, though, I’m not sure how it can be subtracted out first, since the Galaxy is sharing some, but not all, of our motion.
