PDA

View Full Version : Redshift mechanisms and frame transformation



iantresman
2006-Aug-31, 09:15 PM
What is meant by redshift mechanisms? I've read about different types of redshift being associated with different transformation frames.


Redshift type Transformation frame
===================== =======================
Doppler redshift Galilean transformation
Relativistic Doppler Lorentz transformation
Cosmological redshift General relativistic transformation
Gravitational redshift General relativistic transformation

But if I was to discuss redshift mechanisms, what would be inferred? The Redshift type, transformation frame, or something else?

For example, would you say that redshift mechanisms are due to Galilean, Lorentz and General Relativistic transmformation, or described by them?

Regards,
Ian Tresman

Ken G
2006-Sep-01, 04:20 AM
I'm not sure I see the distinction between "described by" and "due to". Is an electric force described by, or due to, an electric field? The only apparent distinction is whether or not there is a cause and effect relationship, or just an identity of some kind, and if that is the sense of "due to" then it is not appropriate-- there is no formal cause and effect relationship between redshift and transformations, in the temporal sense that a cause must precede an effect. But it seems a picky distinction to me. The real point is, no one knows where redshifts come from, but if you want to predict them, you need to follow the necessary transformations. So they are certainly described by the transformations, but it's not that much of a stretch to say they are due to them also.

iantresman
2006-Sep-01, 08:58 AM
If I was to apply the same question to an ambulance's siren as to travels along a street, I would argue that it is more intuitive to say that mechanism behind the Doppler effect is the moving ambulance.

Undoubtedly, there is an associated transformation, but I'd argue that the transformation is not the cause. We don't "see" transformations" causing the Doppler effect. The transformation is just a mathematical abstraction that describes what is going on?

Regards,
Ian Tresman

Ken G
2006-Sep-01, 09:11 AM
Yes, your example points out that it is quite incorrect to associate the normal Doppler shift, say of sound, to a Galilean transformation. That one was just wrong, strike it from your table and alert the source. The Doppler shift of sound works because of motion relative to a medium, along with the statement that the speed is constant in the medium-- it is not a coordinate transformation at all. Or put differently, it is only a Galilean transformation with regard to the motion of the receiver, not the source as in your example. Coordinate transformations are relevant only to mediumless propagation, i.e., to light and to the deBroglie wavelength of particles, because then you are transforming between observers with no absolute reference. That being said, you must now try to find a new example that will establish your point, and you may find that difficult. You can already see what happens to your claim when it is the listener who is moving through the air, and not the ambulance. Where is the cause now?

iantresman
2006-Sep-01, 03:10 PM
Thanks for clarifying that. Let's stick with the Doppler redshift. I would describe the mechanism as being due to motion of the source or observer... and this may be described as a Galilean transformation.

Or perhaps "mechanism" has different meanings according to physicists and mathematicians.

I see "mechanism" as being related to mechanics, something physical. Whereas "transformations" are mathematical abstraction that only exist in textbooks?

Regards,
Ian Tresman

Ken G
2006-Sep-01, 03:29 PM
I see "mechanism" as being related to mechanics, something physical. Whereas "transformations" are mathematical abstraction that only exist in textbooks?


The intersection of the two ideas is that to make a transformation meaningful, you have to have an observer in both frames. But if you have an observer on both sides of the transformation, you can call that "mechanics", since that's all motion really is. A transformation is how you go from the reference frame of one observer to another who is mechanically in motion relative to the first. So it is a mathematical entity, but rooted in physical reality. The transformation is what happens to the observables when you change reference frames, and I don't know what could be more physical than that.

iantresman
2006-Sep-01, 04:52 PM
So if the transformation is "what happens", then isn't it a model, rather than the mechanism (ie what causes) the redshift?

And as an abstract model, there is no information about the mechanism, which is arguably why there have been many mechanisms proposed?

Regards,
Ian Tresman

Ken G
2006-Sep-02, 01:58 AM
So if the transformation is "what happens", then isn't it a model, rather than the mechanism (ie what causes) the redshift?
Nothing "causes" redshifts, in the traditional sense of that word, they are simply observed in well understood and easily predicted situations. For example, the relativistic Doppler shift of light cannot uniquely be said to "happen" at the point of emission, nor at absorption, nor gradually along the way, it depends on your reference frame and how you choose to describe the process. On the other hand, a true "cause and effect" is a clear relationship between events, involving some kind of signal propagating from the cause to the effect, and the cause occurs prior to the effect in all observer frames.

And as an abstract model, there is no information about the mechanism, which is arguably why there have been many mechanisms proposed?
That sounds reasonable to me. The mechanism depends on the manner chosen to describe the effect, the coordinate system if you will. Thus it is not a physical cause and effect relationship.