PDA

View Full Version : Red Shift and the Doppler Effect



Naz
2016-Jun-10, 01:13 PM
Hello, my first post:

I start a course in September: An Introduction to Astrophysics and Cosmology, and I have started reading one of the course texts An Introduction to Galaxies and Cosmology by Jones,Lambourne and Serjeant. I am enjoying it but getting a bit confused by the explanation of red shift. It was described as a phenomena distinct from Doppler.

The red Shift result is simple in its outcome. It's more the explanation and the distinction between red shift and the doppler effect that I do not appreciate. The Doppler Effect is due to Relative motion between source and observer. My text book says that "Redshift is not attributable to the Doppler Effect".

I assume that all that they are getting at here is that the end result is the same, but that Doppler Shift produced by a galaxy moving through space away from us, is different in cause from a distant galaxy moving away from us because space itself is expanding. Same effect just different case?

I am a bit unsure how space itself is created, doesn't that have to be through the motion of an outer envelope of space, or matter but into what? I realise this has probably been done a thousand times.

I have also been looking at some videos by a chap called Michael van Biezen who does a lot of teaching courses for students on Youtube. he has a series on the calculating of distance and there are some videos about how the ancient Greeks calculated the circumference of the Earth, and the diameter of the Moon from really clever observations.

I think I am just going to carry on ploughing through my text book otherwise I will get easily distracted. Thank you very much for reading.

StupendousMan
2016-Jun-10, 07:37 PM
A paper just submitted to arXiv this week discusses some of these issues. I recommend you read it:

On The Relativity of Redshifts: Does Space Really "Expand"? (http://arxiv.org/abs/1605.08634)

Short summary: one can explain all redshifts as expressions of the same underlying phenomenon. My personal take: just accept them and use them as very convenient tools.

Naz
2016-Jun-18, 04:01 PM
A paper just submitted to arXiv this week discusses some of these issues. I recommend you read it:

On The Relativity of Redshifts: Does Space Really "Expand"? (http://arxiv.org/abs/1605.08634)

Short summary: one can explain all redshifts as expressions of the same underlying phenomenon. My personal take: just accept them and use them as very convenient tools.

I read it a couple of times and I think I understood. The red-shift effect is a relativistic effect where the result depends on the frame of reference of the observer. However it raised a new question. Does space expand? Well we know the universe expands and we measure that through the red shift of different galaxies and we have various estimates of the Hubble constant which measures the expansion rate.

I started off with a view that red shifting of stellar objects eg galaxies was two fold.

1) Local movements of galaxies with respect to each other i.e. galaxies travelling at different velocities and

2) Space itself expanding which is the cosmological Red-Shift.

Having read that paper I am troubled by the concept of space expanding. Perhaps it's just simpler than than but two objects separated by empty space have simply moved apart and there is no distinction between point 1 and 2.

I think I'll read the paper again in a little while. I've saved it for reference and will re-read it later and no doubt get more out of it.
Thank you

Ken G
2016-Jun-18, 07:47 PM
However it raised a new question. Does space expand?No one knows, because we cannot do experiments on space. However, this is a very useful picture that cosmologists generally use, so I recommend you also adopt that picture, and realize that it is just a kind of language for talking about what is going on. It is related to a choice of coordinates, when you get much deeper into relativity you will begin to understand the way a choice of coordinates affects our language for talking about what is going on. It is an advanced insight that the language we use is no more unique or absolute than the coordinates we choose to support that language. It sounds to me like you are already asking yourself questions that normally come up much later in most people's education-- if ever!

I started off with a view that red shifting of stellar objects eg galaxies was two fold.

1) Local movements of galaxies with respect to each other i.e. galaxies travelling at different velocities and

2) Space itself expanding which is the cosmological Red-Shift.

Having read that paper I am troubled by the concept of space expanding. Perhaps it's just simpler than than but two objects separated by empty space have simply moved apart and there is no distinction between point 1 and 2.Yes, space expanding is a better picture than motion between the galaxies, but the distinction is rarely significant at the elementary level of description. It really only comes up in general relativity, but it is common to stress that difference anyway. Until you deeply into general relativity, the distinction will probably raise more questions than it answers. In the mean time, just imagine that the galaxies are like raisins in a loaf of raisin bread, and the rising loaf of bread is "space itself." This picture dovetails nicely with the way general relativity works, and also what is known as the "cosmological principle", the idea that everywhere in the universe is doing more or less the same things on large enough scales-- like the loaf of bread. Welcome to the forum.

George
2016-Jun-20, 02:17 PM
It may help -- a slap in the face for me -- to recognize that the observable edge is traveling faster than the speed of light. A Doppler model fails at v > c.

AFJ
2016-Jun-20, 03:15 PM
It may help -- a slap in the face for me -- to recognize that the observable edge is traveling faster than the speed of light. A Doppler model fails at v > c.

Towards us or away from us :)?

George
2016-Jun-20, 03:44 PM
Towards us or away from us :)?Space is accelerating away, due likely to my specious puns.

profloater
2016-Jun-20, 03:52 PM
Space is accelerating away, due likely to my specious puns.
I like that; the specious time continupun has a sense of humour, actually there's loads of evidence for that.

AFJ
2016-Jun-20, 04:53 PM
Space is accelerating away, due likely to my specious puns.

And so the observable edge is coming closer i've recently learned, sort of the room is getting bigger but the walls are closing in.

Ken G
2016-Jun-20, 05:41 PM
It may help -- a slap in the face for me -- to recognize that the observable edge is traveling faster than the speed of light. A Doppler model fails at v > c.Doppler models never fail, you simply don't use the same v. (You don't use the same coordinates either, and that's why we don't use Doppler models-- the necessary coordinates are of little value, compared to the comoving-frame coordinates that support language like "space itself is expanding.")

Ken G
2016-Jun-20, 05:43 PM
I like that; the specious time continupun has a sense of humour, actually there's loads of evidence for that.I think the Hitchhiker's Guide to the Galaxy series is devoted to exploring that evidence...

George
2016-Jun-20, 06:43 PM
Doppler models never fail, you simply don't use the same v. (You don't use the same coordinates either, and that's why we don't use Doppler models-- the necessary coordinates are of little value, compared to the comoving-frame coordinates that support language like "space itself is expanding.") That would be something like a Doppler 2.0 or a relativistic version or something, right? I assume we are only considering the traditional Doppler effect, which is how I first considered the redshift. Regular Doppler makes sense given the emitter is moving at a different velocity than we are, but it fails with very far galactic distances.

George
2016-Jun-20, 06:47 PM
I think the Hitchhiker's Guide to the Galaxy series is devoted to exploring that evidence...

Did you mean exploring or exploding?

Grey
2016-Jun-20, 08:24 PM
That would be something like a Doppler 2.0 or a relativistic version or something, right? I assume we are only considering the traditional Doppler effect, which is how I first considered the redshift. Regular Doppler makes sense given the emitter is moving at a different velocity than we are, but it fails with very far galactic distances.You'd need to use the relativistic version of the equation, but I think most people would still call that the "regular" Doppler effect. And Ken is right that the coordinates get weird if you do this. The recession velocity for distant galaxies does end up less than the speed of light, but the distance becomes infinite. Ned Wright has a nice page (http://www.astro.ucla.edu/~wright/cosmo_02.htm) that shows visually what you get if you switch from "expanding universe" coordinates to "special relativity in static spacetime" coordinates.

AFJ
2016-Jun-21, 10:49 AM
You'd need to use the relativistic version of the equation, but I think most people would still call that the "regular" Doppler effect. And Ken is right that the coordinates get weird if you do this. The recession velocity for distant galaxies does end up less than the speed of light, but the distance becomes infinite. Ned Wright has a nice page (http://www.astro.ucla.edu/~wright/cosmo_02.htm) that shows visually what you get if you switch from "expanding universe" coordinates to "special relativity in static spacetime" coordinates.

Would one expect a difference in the ratio of redshift / absorption lines between the two models?

Would one expect different absorption lines between the two models? For instance if you take the simple example of drawing a 'wave' on a stretchable material, and a vertical line to represent an absorption line. If you now stretch the material to represent expanding spacetime the wave is lenghtenend, but also the vertical line is broadened.

As for a relativistic velocity it can be viewed ( choice of coordinates) ,from our perspective, that the absorption line is already displaced at the exact moment of emission, so would not be expected to be broadened?

Hornblower
2016-Jun-21, 11:21 AM
Would one expect a difference in the ratio of redshift / absorption lines between the two models?

Would one expect different absorption lines between the two models? For instance if you take the simple example of drawing a 'wave' on a stretchable material, and a vertical line to represent an absorption line. If you now stretch the material to represent expanding spacetime the wave is lenghtenend, but also the vertical line is broadened.

As for a relativistic velocity it can be viewed ( choice of coordinates) ,from our perspective, that the absorption line is already displaced at the exact moment of emission, so would not be expected to be broadened?

It is my understanding that all wavelengths are multiplied by the same factor for any given recession rate of the source, whether we model it as motion through space or expanding space. A line that can be broadened already has some intrinsic width, meaning the substance emits or absorbs light over a small range of wavelengths rather than at a single wavelength. If each wavelength is doubled by redshift then the width of the line will be doubled.

Grey
2016-Jun-21, 12:57 PM
Hornblower is right here. The difference between the two diagrams on Ned Wright's page is literally just a change in coordinates. As such, there are no physical differences at all, and no expected differences in measurement, because they are describing exactly the same thing, just using math in different ways. Using special relativity in static spacetime is more difficult and less useful for pretty much all of cosmology (just like it would be more difficult and less useful to work out orbits of planets in the solar system by using the latitude and longitude as measured from a "fixed" Earth), so we don't do it. But we could if we really wanted to, and all measurable quantities would stay exactly the same.

Cougar
2016-Jun-21, 01:20 PM
Ned Wright has a nice page (http://www.astro.ucla.edu/~wright/cosmo_02.htm) that shows visually what you get if you switch from "expanding universe" coordinates to "special relativity in static spacetime" coordinates.

I've seen that before, but certainly worth a re-read. I notice:



If seen at a redshift z, the observed temperature [of a black body - the CMB] will be

Tobs = Tem/(1+z)



We know Tobs = 2.725 K. and z = ~1100, so

the temperature of the CMB when the photons were emitted was right around 3,000 K (or 4,940 F).

Ken G
2016-Jun-21, 01:25 PM
Yes, that's the temperature for hydrogen recombination at the densities present at the time.

AFJ
2016-Jun-21, 03:13 PM
It is my understanding that all wavelengths are multiplied by the same factor for any given recession rate of the source, whether we model it as motion through space or expanding space. A line that can be broadened already has some intrinsic width, meaning the substance emits or absorbs light over a small range of wavelengths rather than at a single wavelength. If each wavelength is doubled by redshift then the width of the line will be doubled.

So an absorption line of a single wavelength will not be broadened by redshift / 'stretching' of the wavelengths, as would happen in the stretch material example given? Why not?

If i erase a very small part of a drawn wave on stretch-material, and then stretch it the erased part will be broader than the initial erasion width.

Grey
2016-Jun-21, 04:39 PM
So an absorption line of a single wavelength will not be broadened by redshift / 'stretching' of the wavelengths, as would happen in the stretch material example given? Why not?

If i erase a very small part of a drawn wave on stretch-material, and then stretch it the erased part will be broader than the initial erasion width.If the line has zero width, it will still have zero width regardless of the redshift mechanism. But of course all real spectral lines have some width (and the details of how that line is being produced may affect that width). Redshift will indeed broaden that line, but it will be the same amount of broadening whether the "mechanism" is expanding space or recession at relativistic velocity.

Let's look at a concrete example. Suppose you're looking at light from some distant object, and you're paying attention to the strong emission line of H\alpha at 656.3 nm. Let's say this line has a width (a typical way of measuring the width of such a line is "full width at half maximum", or how wide the peak is at half its full height when looking at a graph of the spectrum) of 3.0 nm. So that means we're looking at a range from 654.8 nm to 657.8 nm. Now let's redshift this light by a factor of z = 1 (i.e., the wavelength increases by a factor of 1 + z = 2; it doubles). The mechanism doesn't change how redshift changes the wavelength of the light. Now the center of my line will be at 1,312.6 nm, and the range (still looking at FWHM) will be from 1,309.6 nm (that's 2 x 654.8 nm) to 1,315.6 nm (which is 2 x 657.8 nm), giving a width of 6.0 nm. The width of the line has broadened by a factor of 2, exactly the same as the change in the actual wavelength. No matter what causes the redshift, be it an intense gravitational field, recession of the galaxy at high velocity, or expansion of the universe, I'll get the same numbers and observe the same spectrum from this object.

AFJ
2016-Jun-21, 07:36 PM
If the line has zero width, it will still have zero width regardless of the redshift mechanism. But of course all real spectral lines have some width (and the details of how that line is being produced may affect that width). Redshift will indeed broaden that line, but it will be the same amount of broadening whether the "mechanism" is expanding space or recession at relativistic velocity.

Let's look at a concrete example. Suppose you're looking at light from some distant object, and you're paying attention to the strong emission line of H\alpha at 656.3 nm. Let's say this line has a width (a typical way of measuring the width of such a line is "full width at half maximum", or how wide the peak is at half its full height when looking at a graph of the spectrum) of 3.0 nm. So that means we're looking at a range from 654.8 nm to 657.8 nm. Now let's redshift this light by a factor of z = 1 (i.e., the wavelength increases by a factor of 1 + z = 2; it doubles). The mechanism doesn't change how redshift changes the wavelength of the light. Now the center of my line will be at 1,312.6 nm, and the range (still looking at FWHM) will be from 1,309.6 nm (that's 2 x 654.8 nm) to 1,315.6 nm (which is 2 x 657.8 nm), giving a width of 6.0 nm. The width of the line has broadened by a factor of 2, exactly the same as the change in the actual wavelength. No matter what causes the redshift, be it an intense gravitational field, recession of the galaxy at high velocity, or expansion of the universe, I'll get the same numbers and observe the same spectrum from this object.

That is a nice real world example, thanks.

Are there any known observations and experiments of redshifted absorption - emission lines / line widths of high speed subjects on earth?

I can't seem to find any, should be almost the same principle as this one example i did find from more than a 100 years ago;

http://adsabs.harvard.edu/abs/1901ApJ....13...15B

Naz
2016-Jun-22, 05:39 PM
No one knows, because we cannot do experiments on space. However, this is a very useful picture that cosmologists generally use, so I recommend you also adopt that picture, and realize that it is just a kind of language for talking about what is going on. It is related to a choice of coordinates, when you get much deeper into relativity you will begin to understand the way a choice of coordinates affects our language for talking about what is going on. It is an advanced insight that the language we use is no more unique or absolute than the coordinates we choose to support that language. It sounds to me like you are already asking yourself questions that normally come up much later in most people's education-- if ever!Yes, space expanding is a better picture than motion between the galaxies, but the distinction is rarely significant at the elementary level of description. It really only comes up in general relativity, but it is common to stress that difference anyway. Until you deeply into general relativity, the distinction will probably raise more questions than it answers. In the mean time, just imagine that the galaxies are like raisins in a loaf of raisin bread, and the rising loaf of bread is "space itself." This picture dovetails nicely with the way general relativity works, and also what is known as the "cosmological principle", the idea that everywhere in the universe is doing more or less the same things on large enough scales-- like the loaf of bread. Welcome to the forum.
Thanks for replying. I came across a section about co-moving coordinates and use of a scale factor to describe the physical separation of points in space.

I don't think space can expand. By space presumably we mean the vacuum between particles where no matter exists. The particles themselves cannot expand. They are in the lowest energy states possible. So if matter cannot expand, and space is the absence of matter then what we have left is the boundaries of the universe expanding and the contents flying apart for now?

I remember that the rate of expansion of two galaxies in space is proportional to their distance apart, and the scale factor is increasing. At the time I read that I thought well that makes sense if space is expanding, but it must be the case that in general matter is flying apart from each other under the influence of Dark energy and because that Force is ever present acceleration continues and hence the Hubble constant grows.

StupendousMan
2016-Jun-22, 11:30 PM
That is a nice real world example, thanks.

Are there any known observations and experiments of redshifted absorption - emission lines / line widths of high speed subjects on earth?

I can't seem to find any, should be almost the same principle as this one example i did find from more than a 100 years ago;

http://adsabs.harvard.edu/abs/1901ApJ....13...15B

If you go halfway through my lectures notes here

http://spiff.rit.edu/classes/phys314/lectures/doppler/doppler.html

you'll find a description of an experiment done in 1938 at Bell Labs which succeeded in measuring the relativistic Doppler effect for hydrogen ions.

AFJ
2016-Jun-23, 10:50 AM
If you go halfway through my lectures notes here

http://spiff.rit.edu/classes/phys314/lectures/doppler/doppler.html

you'll find a description of an experiment done in 1938 at Bell Labs which succeeded in measuring the relativistic Doppler effect for hydrogen ions.

That's cool! appreciate it, i could not find anything thanks for sharing. Does seem like 'old' experiments, maybe because they were succesfull and so precise reproducing them isn't necessary anymore, nothing on line widths though.

EDIT amazing within 1% of the relativistic prediction! That's awfully precise engineering :)