Is the Cosmic Microwave Background an Absolute Frame of Reference?

[mrcotton]

I heard sometime ago that the cosmic microwave background may be a candidate for an absolute frame of reference. Did this idea ever get any credence?

 

[Dale]

The cosmic microwave background radiation doesn't even define a single inertial frame, let alone an absolute one.

 

[tom.stoer]

In a sense it does.

Suppose CMB has no fluctuations. Then the CMB defines a unique frame by demanding that all observed multipoles (except for the monopole) vanish.

 

[Dale]

Then the CMB defines a unique frame by demanding that all observed multipoles (except for the monopole) vanish.

That isn't a unique frame. That is a different frame for every location in the universe, all moving wrt each other.

That is my point, the CMB doesn't even define a unique frame, let alone an absolute frame. It is simply an idea that is "dead on arrival".

 

[tom.stoer]

OK, I agree, the CMB defines a global principle to uniquely fix a family of local reference frames.

 

[DiracPool]

We touched on this subject some time ago in this thread:

https://www.physicsforums.com/showthread.php?t=664897&highlight=moving+relative+cosmic

Look around starting at about post #15 for discussion and references

From the CMB data it is seen that our local group of galaxies (the galactic cluster that includes the Solar System's Milky Way Galaxy) appears to be moving at 627±22 km/s relative to the reference frame of the CMB (also called the CMB rest frame, or the frame of reference in which there is no motion through the CMB) in the direction of galactic longitude l = 276±3°, b = 30±3°.[66] This motion results in an anisotropy of the data (CMB appearing slightly warmer in the direction of movement than in the opposite direction).[67] The standard interpretation of this temperature variation is a simple velocity red shift and blue shift due to motion relative to the CMB

So, it seems as if they are implying an absolute reference frame, or one global static frame that sits relative to our local group at least, by the detection of a doppler shift in these galaxies. But what constitues that rest frame?

Suppose CMB has no fluctuations. Then the CMB defines a unique frame by demanding that all observed multipoles (except for the monopole) vanish.

tom.stoer, can you eaborate on this? I'm not sure I understand what multipoles are.

Is the idea that, as the universe expands, it leaves in its wake a "signature" of its expansion? This being the CMB radiation. Say the analogy being like you blow up a balloon, and at each infinitesimal expansion in the radius you leave a marker of where the surface of the sphere just was, such that the volume of the balloon has a record of these, in a sense, absolute intertial reference points that continue to accumulate as the radius of the balloon expands?

How would these reference points be manifested in the case of our universe and the CMBR? Is this where the multipoles come in?

 

[pervect]

We touched on this subject some time ago in this thread:So, it seems as if they are implying an absolute reference frame, or one global static frame that sits relative to our local group at least, by the detection of a doppler shift in these galaxies. But what constitues that rest frame?

This has already been answered once:

That isn't a unique frame. That is a different frame for every location in the universe, all moving wrt each other.

That is my point, the CMB doesn't even define a unique frame, let alone an absolute frame. It is simply an idea that is "dead on arrival".

I think that the answer is already pretty clear, but let me repeat it just in case.

At any given event (or point) in space-time, the CMB does single out one particular local frame. This local frame, is still local, it's not a "global frame".

The reason why it's not a global frame is the fact that a different frame is singled out at different points - in particular, the local frames singled out in this manner are all moving with respect to one another (as one varies the selection point).

In some sense, CMB defines a preferred coordinate system. Objects at rest in the CMB define cosmological time, and cosmological time defines a preferred set of spatial slices.

However, the resulting set of coordinates is not a "frame" of reference in the sense that is meant by SR. One needs to use GR with these cosmological coordinates. The metric that is associated with (one can even say the metric that defines) these coordinates is the Friedmann–Lemaître–Robertson–Walker (FLRW) metric.

The FLRW metric is common, and useful, but it's not a "frame" in the technical sense used by SR, you can't deal with it by the methods of SR. The fundamental reason behind this is that the metric has underlying curvature (at least in any universe with matter, and without matter it's rather hard to imagine how one would verify the cosmological principle that's underlying the whole discussion).

 

[tom.stoer]

tom.stoer, can you eaborate on this? I'm not sure I understand what multipoles are.

First of all let me say that there is no rest frame w.r.t. CMB b/c CMB is light-like.

The idea is the following: in a homogeneous and isotropic universe, CMB should also be homogeneous and isotropic. But any motion (w.r.t. to a locally defined sphere from which the CMB is received) may create anisotropies of the CMB on the celestial sphere, which can be expanded in multipoles (this is what they are doing when analyzing the Planck data). The zeroth multipole = the monopole is nothing else but the averaged CMB. The second multipole = the dipole measures nothing else but the motion w.r.t. CMB-frame (the above mentioned sphere). So in a sense one can always try to find a frame where the dipole vanishes, and interpret this as a unique reference frame (but as said this has to be done locally, so there is not just one such frame, but a family of such frames, one for each spacetime point - neglecting trivial rotations). Higher multipoles i.e. fluctuations will vanish due to isotropy.