PDA

View Full Version : Redshifting - energy goes somewhere



Professor Illwill
2007-Feb-09, 01:11 PM
Maybe I'm too lazy. Given time I suspect I could figure this one out, but I'm impatient and it's really bugging me.

Body A & Body be are moving apart. Pick a rate. I don't care about gravitational effects or any of the other redshifting - just straight-up doppler shifting. A expends energy to create light, and when it gets to B there's less energy available simply due to motion. Doesn't matter if it's absorbed, reflected, whatever. There's always less energy at B, and I don't see any way to recover it.

The only mechanism that came to my mind was if the amplitude of the light increased in proportion to the redshift (I'm not a photon man). This seems absurd. I just can't figure out what happens to the energy so that it will be conserved. At first I thought maybe the same amount of energy arrives, only at a slower rate; but this is also incorrect.

Now I'll grant that many materials absorb IR better than UV, so the exchange might seem to work out "better" for B. Don't get distracted with that, please. The fact is that no matter what happens at B, the light's going to be packing less punch when it arrives.

It's hard to word a search to screen out 2000 unwanted results. I found some nonsense about the energy being transferred to the aether field, but that can't be right. Photonists won't buy it for an instant, and I find it most unsatisfactory. The aether isn't just a catch-all waste dump for things we can't explain. If it can absorb energy it's gotta give it back (some odd version of Steady-State?). Most importantly, the transfer of energy to the aether under these rules would require it to be postponed until the light reaches B. Otherwise some C object moving toward A could intervene and the light would have to suddenly, magically increase its energy content because C is blue-shifted vs. A. Nonsense!

Note: photon answers are fully acceptable, so long as we find this energy. There's gotta be a "pat" answer for this one after all this time.

Kaptain K
2007-Feb-09, 04:53 PM
The total energy does not change, but the energy per unit time does - in direct proportion to the doppler shift. Let's say that object A emits a one second pulse with a wavelength of 350 nm (soft UV) and, because of doppler shift, object B receives a two second pulse with a wavelength of 700 nm (near IR). The energy per unit time is halved, but the total is unchanged.

Squashed
2007-Feb-09, 04:59 PM
The total energy does not change, but the energy per unit time does - in direct proportion to the doppler shift. Let's say that object A emits a one second pulse with a wavelength of 350 nm (soft UV) and, because of doppler shift, object B receives a two second pulse with a wavelength of 700 nm (near IR). The energy per unit time is halved, but the total is unchanged.

Isn't this analogy similar to saying that energy transfers continually slow as time (or the age of the universe) progresses?

When the photon was emitted time flowed slower and since the photon does not change then since time is flowing faster, now, the reception occurs slower.

Squashed
2007-Feb-09, 05:01 PM
...a photon man...

Sounds like a good "screen name".

Hi, my name is PhotonMan.

Kwalish Kid
2007-Feb-09, 05:35 PM
Isn't this analogy similar to saying that energy transfers continually slow as time (or the age of the universe) progresses?

When the photon was emitted time flowed slower and since the photon does not change then since time is flowing faster, now, the reception occurs slower.
The expansion means that from the time frame of the receiver, the emission was slower than it would be in the emitter's time frame. (That's a little loose there, but hopefully you get the point.) And, of course, vice versa.

George
2007-Feb-09, 08:26 PM
I still like the snowball - bus analgoy. Throw a snowball at a still bus and it smacks it pretty hard. But when the bus is moving, the impact of the snowball is less.

The difference between the two impacts is really the difference in the relative energy, though, looking deeper in this analogy, the snowball's energy is decreased because of the difference in actual velocites per KE. For a photon, it is a difference in the relativistic mass of the photon. The faster the bus is traveling, the less will be the relativistic mass of the object, and the weaker the impact, meaning more redshift for a photon.

However, wait for others to comment before swallowing this view.

Ken G
2007-Feb-09, 09:28 PM
Note: photon answers are fully acceptable, so long as we find this energy. There's gotta be a "pat" answer for this one after all this time.There is:

I still like the snowball - bus analgoy. Throw a snowball at a still bus and it smacks it pretty hard. But when the bus is moving, the impact of the snowball is less. Yes, this is very much the view you should take if you want the photon picture, which is what seems to have been favored in the OP. But here's the missing element-- the "less impact" that George is talking about is the energy extracted in the frame of the absorber, whereas energy conservation requires that you stay in the frame of the emitter. The first thing to realize is that the amount of energy that is being transferred is different in different reference frames, so step 1 is to pick a frame and stick with it, if you want energy to be conserved (that this is not done in GR is why you don't expect conservation of energy in GR). So let's pick a frame where the emitter is stationary and the absorber is moving.

The next key step is to recognize that the momentum gained by the absorber is the same as that given up by the emitter, by conservation of momentum. But here's where the magic comes in-- when you take a tiny amount of momentum from something stationary, there is no loss of energy from the stationary source, other than the energy that is going into the "thing" itself (i.e., the energy of the photon). But when you transfer momentum to something moving, there is energy gained by the motion of the moving object. Thus to make energy balance, you actually must have less energy dissipated (or "extracted") by the moving object in the moving object's frame (i.e. it must "think" it is absorbing a lower energy photon), because that frame is oblivious to the energy that is seen as going into its motion in the first frame! Thus the redshift of the photon does not create a mysterious absence of conservation of energy, rather it is required in order to conserve energy in the fixed choice of reference frame. It is only when you "change horses" in midstream and shift reference frames that you think you are seeing a lack of conservation of energy. Don't change frames when you do the energy budget, is the bottom line.

Sam5
2007-Feb-09, 09:42 PM
Maybe I'm too lazy. Given time I suspect I could figure this one out, but I'm impatient and it's really bugging me.

Body A & Body be are moving apart. Pick a rate. I don't care about gravitational effects or any of the other redshifting - just straight-up doppler shifting. A expends energy to create light, and when it gets to B there's less energy available simply due to motion. Doesn't matter if it's absorbed, reflected, whatever. There's always less energy at B, and I don't see any way to recover it.

It's called "less energy" when photons redshift (less energetic photons), however, what happens is that the photons are stretched out in time or space or both. So the same overall energy that is emitted is received but over a longer time period. With a blueshift they are received in a shorter period of time.

Ken G
2007-Feb-09, 09:50 PM
It's called "less energy" when photons redshift (less energetic photons), however, what happens is that the photons are stretched out in time or space or both. So the same overall energy that is emitted is received but over a longer time period.
Several posts have already made this misleading claim. I'm afraid you'll have to see posts #6 and #7 to get the straight story.

George
2007-Feb-09, 11:20 PM
But here's the missing element-- the "less impact" that George is talking about is the energy extracted in the frame of the absorber,... Yes, thanks for clarifiying this important point. We are all on the bus, now, and we will see where it goes. :)


The next key step is to recognize that the momentum gained by the absorber is the same as that given up by the emitter, by conservation of momentum. But here's where the magic comes in-- when you take a tiny amount of momentum from something stationary, there is no loss of energy from the stationary source, other than the energy that is going into the "thing" itself (i.e., the energy of the photon).
This makes sense if it is from the reference frame of the emitter (I'm on the bus, still)?


But when you transfer momentum to something moving, there is energy gained by the motion of the moving object. Thus to make energy balance, you actually must have less energy dissipated (or "extracted") by the moving object in the moving object's frame (i.e. it must "think" it is absorbing a lower energy photon), because that frame is oblivious to the energy that is seen as going into its motion in the first frame! It is almost, or perhaps it is, the same as in the case where a rock is thrown upward and just barely reaches and hits the bottom of a steel bridge, it makes much less of a noise than if the bridge were much lower. This is because there is so much energy required from the thrower to overcome the potential energy difference which is taken away from the initial kinetic energy in the case of the highly elevated bridge. [I seem to be throwing stuff today. :)]

Sam5
2007-Feb-10, 12:08 AM
The total energy does not change, but the energy per unit time does - in direct proportion to the doppler shift. Let's say that object A emits a one second pulse with a wavelength of 350 nm (soft UV) and, because of doppler shift, object B receives a two second pulse with a wavelength of 700 nm (near IR). The energy per unit time is halved, but the total is unchanged.


Correct.

Ken G
2007-Feb-10, 01:49 AM
No Sam5, Kaptain K is not correct, photon energy doesn't work that way, as I'm sure he'll quickly recognize when he sees posts #6 and #7. You, on the other hand, will be slower, but remember this is Q&A, so our job is actually to present correct answers to the best of our ability. We all make mistakes, some are just quicker to recognize it.

Sam5
2007-Feb-10, 02:57 AM
No Sam5, Kaptain K is not correct,

If you have a problem with Kaptain K's post, then discuss it with him, not me.

Kaptain K
2007-Feb-10, 03:06 AM
Hmm...I'm speaking in the "wave realm". You move to the "particle (photon) realm" and tell me I'm wrong!?

Professor Illwill
2007-Feb-10, 10:43 AM
Thank you all. I'm comfortable that I understand this now.

The snowball really helped (that's why I try to always blue-shift my snowball!).

It's easier to keep frames straight there, since when the bus stops it all equals out quite obviously. The snow must either stick to the bus or bounce off. (some of both in my experience). That which sticks adds to the mass of the bus & the brakes convert more momentum to heat when the bus stops. That which bounces off transfers part of its momentum to the bus and retains part. Easy.

I'm "getting it" with regards to light as well, although I'm not confident I could explain it to another yet. I at least understand it myself.

I don't think the "wave realm" answer can be incorrect, since all waves are subject to the doppler effect; not just the ones to which we ascribe particle properties. It's a lot like what I was thinking had to be wrong. Too many times I've heard "red-shifted light loses energy". I suppose I took it at face value.

Ken G
2007-Feb-10, 03:03 PM
Hmm...I'm speaking in the "wave realm". You move to the "particle (photon) realm" and tell me I'm wrong!?

You're just as wrong in the wave realm, it's not a "realm" issue (you know that, it's a measurable!). The issue is that you have improperly added up the energy. See post #7. (You can translate that into the wave realm, though it's a bit more effort-- you'll need to note that the energy flux of the wave does not scale in linear proportion to the frequency, as you assumed.)

Ken G
2007-Feb-10, 03:04 PM
If you have a problem with Kaptain K's post, then discuss it with him, not me.
You are the one who reposted his and said "correct" even after I explained why it was wrong. I'm acting on the assumption that Professor Illwill does not want to get confused by contradicting answers, so it behooves us all to get our story straight. But the over-arching point of interest here is that there is often subtle physics hiding behind even the simplest sounding questions, so keep on asking, Professor!

George
2007-Feb-10, 07:11 PM
This thought, gedakenexp., may only complicate it but I am curious what answer y'all might have...

From the reference frame of an emitter, E, in galaxy A, a green pulse of laser light is emitted. After several billion years, the light path is bent by several perfectly situated galaxies that do not alter the light's net energy relative to galaxy A. Is the returning light still green as seen by E?

I still like the Doppler view which says, I think, the light is green. However, if the waves are streeeetched by expansion, as some suggest, then it will redshift.

[Or should this be a separate thread?]

satori
2007-Feb-10, 08:00 PM
I see the photon redshifted due to space expansion going on whilst the thing is doing its loop.
(concerning the original question of this damned DrIIIwill, Iwould have come up with the same exp. as KenG. Photon reflected back redshift revised!)
Ah yes, back to the first point. It looks a little as if space expansion does away with energy or takes its drive from this... but of course that is naive(?)...

Sam5
2007-Feb-10, 08:27 PM
This thought, gedakenexp., may only complicate it but I am curious what answer y'all might have...

From the reference frame of an emitter, E, in galaxy A, a green pulse of laser light is emitted. After several billion years, the light path is bent by several perfectly situated galaxies that do not alter the light's net energy relative to galaxy A. Is the returning light still green as seen by E?

I still like the Doppler view which says, I think, the light is green. However, if the waves are streeeetched by expansion, as some suggest, then it will redshift.

[Or should this be a separate thread?]

I think that is a different topic. Here is some more information about the Doppler Effect stretching out the total energy of light in time and space when redshifts occur:

See this:

"At a redshift of 1.7, three and a half weeks in our frame of reference is only about nine days of elapsed time for the supernova itself."
http://www.lbl.gov/Science-Articles/Archive/oldest-1a-supernova.html

See also:

“Time dilation means that the rate at which photons arrive is reduced by a further factor of (1+z)”
http://72.14.253.104/search?q=cache:OTAuUdRgC7MJ:rocinante.colorado.edu/~pja/astr3830/lecture35.pdf+1a+supernova+time+dilation&hl=en&ct=clnk&cd=34&gl=us


This is an example of the light being stretched out in space, reduced in “energy” when it reaches the earth, AND taking a longer time to reach the earth because the whole “energy” of the light is stretched out in space so that we receive it over a longer period of time.

For further information on the Doppler Effects, see: Christian Doppler, “Uber das farbige Licht der Doppelsterne und einiger anderer Gestirne des Himmels,” Abhandlungen der Königliche Böhmischen Gesellschaft der Wissenschaften, Series V, Vol. II, Prague, 1842. Also published in ABHANDLUNGEN, by Christian Doppler, a collection of 11 important Doppler papers (in German), edited by H.A. Lorentz, published by Wilhelm Engelmann, Leipzig, 1907.

George
2007-Feb-10, 09:24 PM
I think that is a different topic. Not so much different as deeper than intended, I suppose. If a more pure Doppler model is correct, the energy doesn't "go somewhere".


See this:

"At a redshift of 1.7, three and a half weeks in our frame of reference is only about nine days of elapsed time for the supernova itself."
http://www.lbl.gov/Science-Articles/Archive/oldest-1a-supernova.html Nice, but if the light of 1997a returned to itself unaltered at the turns, would it be the same wavelength? They show supporting evidence that we are accelerating, so our relative velocity has increased with respect to the SN. Doppler says it will redshift in accordance with this veolcity. It does not answer the question, as far as I can see.


This is an example of the light being stretched out in space, reduced in “energy” when it reaches the earth, AND taking a longer time to reach the earth because the whole “energy” of the light is stretched out in space so that we receive it over a longer period of time.
I doubt it, though at the risk of looking impish. But that does seem to be their view...

This is quite intuitive - wavelength of radiation expands along with expansion of the Universe. This is clearly wrong - it is not intuitive, at least for me. :) The idea that a healthy photon becomes stretched under the influence of a immeasurably small localized force behind expansion just seems wrong. Have atoms grown in the last 13 billion years?


For further information on the Doppler Effects, see: Christian Doppler, “Uber das farbige Licht der Doppelsterne und einiger anderer Gestirne des Himmels,” Abhandlungen der Königliche Böhmischen Gesellschaft der Wissenschaften, Series V, Vol. II, Prague, 1842. Also published in ABHANDLUNGEN, by Christian Doppler, a collection of 11 important Doppler papers (in German), edited by H.A. Lorentz, published by Wilhelm Engelmann, Leipzig, 1907.
Das bin nicht for one who can not read Deutch. :)

Sam5
2007-Feb-10, 11:01 PM
Das bin nicht for one who can not read Deutch. :)

Lol. I am a jelly doughnut. :)

Ken G
2007-Feb-11, 03:07 AM
From the reference frame of an emitter, E, in galaxy A, a green pulse of laser light is emitted. After several billion years, the light path is bent by several perfectly situated galaxies that do not alter the light's net energy relative to galaxy A. Is the returning light still green as seen by E? The question is ill posed, because it's not clear what constraint you are imposing when you say "the light path is bent by galaxies that do not alter the light's net energy". That sounds like you are saying the action of those galaxies is to deliver a returned pulse with no redshift, which is possible in principle but not for any galaxies we know of. So probably what you mean is, those galaxies are behaving "normally" and aren't doing anything special beyond turning the light around, in which case the expansion of the universe would certainly render the light redshifted. But one could certainly say that the redshift was "due to" those galaxies, or one could say that it isn't-- the "reasons" for redshifts are rarely unique.

I still like the Doppler view which says, I think, the light is green. Again, as you have set up the question, you may get the answer you specify, but you won't find galaxies that can do that a billion light years apart in this universe.

astromark
2007-Feb-11, 03:35 AM
Can I play? . . . Good on you Ken for your explanation once again helps me see this clearly. Not daring for a moment to sagest that any other response is ill founded, just different. I would like to pose a question and then be judged as stupid. As light travels the great distances across the universe to reach the optic equipment receiving said photons. Does it not need to spread its self thinner over distance. In order to cover a 360 deg output it must diminish in energy as distance is covered. What am I missing here? Is it not obvious that given mathematical formulas the drop in magnitude is in direct relation to distance covered. Is this even the same subject? Thank you Ken.

Sam5
2007-Feb-11, 04:21 AM
Can I play? . . . Good on you Ken for your explanation once again helps me see this clearly. Not daring for a moment to sagest that any other response is ill founded, just different. I would like to pose a question and then be judged as stupid. As light travels the great distances across the universe to reach the optic equipment receiving said photons. Does it not need to spread its self thinner over distance. In order to cover a 360 deg output it must diminish in energy as distance is covered. What am I missing here? Is it not obvious that given mathematical formulas the drop in magnitude is in direct relation to distance covered. Is this even the same subject? Thank you Ken.

That’s a very good question.

The answer depends on whether you think photons are long skinny things like wiggling arrows or spherical bubbles. Do photons move "in straight lines" or "expanding bubbles"? If they are "bubbles", expanding "spherically", then how can they form a "beam" and a "ray"?

triclon
2007-Feb-11, 04:31 AM
It's the same amount of waves hitting you right? It just takes more time for it to happen since they're getting streatched out when redshifted. Like if you dumped a whole bucket of water out all at once, as opposed to just tilting it slightly and letting the water slowly drain out. Still the same amount of water but takes longer for it to all fall out the bucket.

Sam5
2007-Feb-11, 05:07 AM
It's the same amount of waves hitting you right? It just takes more time for it to happen since they're getting streatched out when redshifted. Like if you dumped a whole bucket of water out all at once, as opposed to just tilting it slightly and letting the water slowly drain out. Still the same amount of water but takes longer for it to all fall out the bucket.

Generally speaking, that's correct with Doppler redshifts.

Ken G
2007-Feb-11, 05:23 AM
In order to cover a 360 deg output it must diminish in energy as distance is covered. What am I missing here?You're not missing anything, that effect is present as well, just wasn't at issue in the above questions because it doesn't relate to redshift, just dimming.

Ken G
2007-Feb-11, 05:27 AM
It's the same amount of waves hitting you right? It depends on what you mean by the "same amount". If you mean, the same number of wave crests and troughs, in a classical description, then yes, that just gets spread over more time. But if you mean the same total energy seen by the absorber, no, that is what is not the same.
Like if you dumped a whole bucket of water out all at once, as opposed to just tilting it slightly and letting the water slowly drain out.
Not if the amount of water is your analogy for energy-- then the moving absorber would perceive less total water.

sirius0
2007-Feb-11, 10:08 AM
Welcolme and I love your login name.

A thread similar to this that started a whole learning curve (with many of the same stars you have answering you now) is below.

http://www.bautforum.com/showthread.php?t=48114

and this one I started as a stumbling response to concerns I had about time being used with photons.

http://www.bautforum.com/showthread.php?t=50189


I am a little lost as I can't at this ponit think of a situation where wave theory demonstrates the energy that is your initial concern. I can't think of where we have an interaction between waves and matter where energy is imparted to the matter that doesn't require jumping into photon theory. maybe induced current with radio in an antena perhaps. Of course I also think it hard to think doppler in terms of photons

Professor Illwill
2007-Feb-11, 10:18 AM
Can I play? . . . Good on you Ken for your explanation once again helps me see this clearly. Not daring for a moment to sagest that any other response is ill founded, just different. I would like to pose a question and then be judged as stupid. As light travels the great distances across the universe to reach the optic equipment receiving said photons. Does it not need to spread its self thinner over distance. In order to cover a 360 deg output it must diminish in energy as distance is covered. What am I missing here? Is it not obvious that given mathematical formulas the drop in magnitude is in direct relation to distance covered. Is this even the same subject? Thank you Ken.

That's a fine question. It's one reason I'm not a photon man: It's simply inconceivable for any group of particles to disperse out to infinity without gaps.

I noticed this a while back. I found the flips and twists one must do to get around the problem. I reviewed the properties of light to see if it (the particle part) was needed. And hacked it clean off with Occam's Razor.

I didn't start this thread to start an argument, so I'll stop with that summary. PM me if you're interested in non-particulate light.

Now for a comment on the original topic:

I not only understand this COE example much better than I did before; I see a new reason why it must be so/how it could easily be tested. If there were any net change in energy due to doppler effects, we could blue-shift 2 counter-rotating objects and manufacture all kinds of free energy.

Ken G
2007-Feb-11, 01:10 PM
I can't think of where we have an interaction between waves and matter where energy is imparted to the matter that doesn't require jumping into photon theory. maybe induced current with radio in an antena perhaps.
Yes, that would be an excellent situation to consider in the wave picture, but your basic point is well taken-- it is easier to imagine what a Doppler shift is in the wave picture, but the photon picture (George's "snowball") makes the energy transfer clearer. Rest assured you can do either in either picture, and the answer is the same in terms of the fractional reduction in energy that the absorber perceives as being received compared to what the emitter perceives as being sent out. If you are algebraically inclined, you may be able to see this by noting that the energy flux in a light wave of fixed amplitude scales as the square of the frequency, not the frequency, and that's the source of the energy difference in the wave picture.

Ken G
2007-Feb-11, 01:13 PM
I reviewed the properties of light to see if it (the particle part) was needed. And hacked it clean off with Occam's Razor.
Just make sure you're not hacking off the photoelectric effect, and Einstein's Nobel prize, in the process.

If there were any net change in energy due to doppler effects, we could blue-shift 2 counter-rotating objects and manufacture all kinds of free energy.
There is a net change in energy due to Doppler effects, so your analysis of counter-rotating objects is flawed, which is not that surprising since there are many subtle issues that need to be handled properly in such an analysis.

Jeff Root
2007-Feb-11, 05:39 PM
...I'm not a photon man: It's simply inconceivable for any
group of particles to disperse out to infinity without gaps.
What gave you the idea that there aren't gaps between photons?
The dimmer the light, the farther-between the photons are.
A light source emits a definite, finite number of photons per unit
time in any band of the spectrum (such as "visual"). When the
incident light is sufficiently dim, the photons falling on a CCD or
photomultiplier can be detected individually.

-- Jeff, in Minneapolis

George
2007-Feb-11, 09:28 PM
The question is ill posed, because it's not clear what constraint you are imposing when you say "the light path is bent by galaxies that do not alter the light's net energy".
Good point and I muddied-up the gedankenexperiment unnecessarily as there is much easier way to cause the light to return back to the emitter - it is called a mirrror. :wall:

So, send out a green laser to be reflected from a mirror that is several billion lightyears distant from emitter E, then wait the for the return light and tell me the color. The mirror must exhibit no proper motion relative to Galaxy A using the original coordinate system of Galaxy A at the time of emission. This is to eliminate, I trust, the motion of the mirror from affecting the wavelength.

Ken G
2007-Feb-12, 01:35 AM
The mirror must exhibit no proper motion relative to Galaxy A using the original coordinate system of Galaxy A at the time of emission. This is to eliminate, I trust, the motion of the mirror from affecting the wavelength.
It's still ill posed, because I don't know what you mean by the "original coordinate system of Galaxy A". There are many possible such coordinate systems. I think what you mean is that the mirror is supposed to be sitting in a real galaxy somewhere a billion light years away, and the mirror isn't doing anything "unusual". Under those circumstances, the green light comes back a lot redder! The interpretation can be monkeyed with by the coordinates chosen to analyze it.

Professor Illwill
2007-Feb-12, 02:26 AM
Welcolme and I love your login name.

A thread similar to this that started a whole learning curve (with many of the same stars you have answering you now) is below.

http://www.bautforum.com/showthread.php?t=48114

and this one I started as a stumbling response to concerns I had about time being used with photons.

http://www.bautforum.com/showthread.php?t=50189


I am a little lost as I can't at this ponit think of a situation where wave theory demonstrates the energy that is your initial concern. I can't think of where we have an interaction between waves and matter where energy is imparted to the matter that doesn't require jumping into photon theory. maybe induced current with radio in an antena perhaps. Of course I also think it hard to think doppler in terms of photons

Thanks for the links too.

Just when I thought I had a firm grip :doh:

It's mighty hard to resist the temptation to hop to a different frame in some of those scenarios, I'll tell you.

I will get it. I will get it.

George
2007-Feb-12, 02:31 AM
It's still ill posed, because I don't know what you mean by the "original coordinate system of Galaxy A". There are many possible such coordinate systems. I am trying to word it so the affect expansion has on the relatvie velocity of the mirror is nullified with respect to the emitter. I don't want the mirror's motion to be a cause of redshifting.


I think what you mean is that the mirror is supposed to be sitting in a real galaxy somewhere a billion light years away, and the mirror isn't doing anything "unusual". Under those circumstances, the green light comes back a lot redder! The interpretation can be monkeyed with by the coordinates chosen to analyze it. Yes, but I hope you see the coordinates I am trying to establish. If so, it should return green.

Ken G
2007-Feb-12, 07:43 AM
I am trying to word it so the affect expansion has on the relatvie velocity of the mirror is nullified with respect to the emitter. I don't want the mirror's motion to be a cause of redshifting.These statements are not necessarily compatible-- I would have said that you will need to inject a dramatic amount of unusual motion into that mirror to "nullify" the effect of the expansion. Thus it would be the unusual motion of the mirror that would be needed to avoid the redshifting. Perhaps what is causing confusion is that there is no absolute standard to say if the mirror is moving or not, all we can say is that its motion is either like a typical galaxy, or very much not like that.


Yes, but I hope you see the coordinates I am trying to establish. If so, it should return green.
You can get it to return green if you set up the problem that way, but you cannot get that result to emerge naturally using any mirrors that you will be likely to find without a starship attached to them. Any real mirror you could find lying around in that distant galaxy is going to give you a redder result.

Ithildin
2007-Feb-12, 08:15 AM
Hi! I didn't quite follow the previous discussion, but can answer your initial question Professor Illwill. Energy is not lost from point A to point B, because if one considers only the Doppler Shift then the wavelength reaching point B is exactly the same wavelength that was emitted. Energy of a photon relates directly to wavelength (by E = hc/w, where w is wavelength).

"The only mechanism that came to my mind was if the amplitude of the light increased in proportion to the redshift" [from the original post]
You are correct that this would be wierd, and in addition this would not make up for a lack of energy since the amplitude of a photon has nothing to do with its wavelenth (or the color perceived, assuming one was talking about light in the visible spectra).

I hope this helps?

Ken G
2007-Feb-12, 08:36 AM
Hi! I didn't quite follow the previous discussion, but can answer your initial question Professor Illwill. Energy is not lost from point A to point B, because if one considers only the Doppler Shift then the wavelength reaching point B is exactly the same wavelength that was emitted. Welcome to the forum Ithildin (and I love your icon), but I'm afraid your physics needs some patching up. There is no energy lost (where it goes is explained in post #7), but the absorbed wavelength is quite certainly different than the emitted one, I have no idea what you mean by the "Doppler Shift" were that not so.

Ithildin
2007-Feb-12, 08:56 AM
Welcome to the forum Ithildin (and I love your icon), but I'm afraid your physics needs some patching up. There is no energy lost (where it goes is explained in post #7), but the absorbed wavelength is quite certainly different than the emitted one, I have no idea what you mean by the "Doppler Shift" were that not so.

Thankyou for the welcome :)
I understand what you are saying, and am sorry that my explaination wasn't clear; its worded horribly. I'm finding it difficult to explain without math or pictures, haha.

George
2007-Feb-12, 02:30 PM
Perhaps what is causing confusion is that there is no absolute standard to say if the mirror is moving or not, all we can say is that its motion is either like a typical galaxy, or very much not like that.
Yes. So let me improve the gedankenexperiment, maybe. We place the mirror at the end of a long pole and extend it out 3 or 4 billion lightyears out. Of course, we avoid gravity wells and we place it along an isotherm. This should give us the null relative velocity necessary. If galaxies are essentially unaffected by expansion, than I assume the pole's tensile strength is even less affected. However, if light is affected by the expansion by streeetching the wave, as some say, then it will be affected and return red; yet I say green.


You can get it to return green if you set up the problem that way, but you cannot get that result to emerge naturally using any mirrors that you will be likely to find without a starship attached to them. A starship(!?), no need to get elaborate with the props. ;)

George
2007-Feb-12, 02:30 PM
Perhaps what is causing confusion is that there is no absolute standard to say if the mirror is moving or not, all we can say is that its motion is either like a typical galaxy, or very much not like that.
Yes. So let me improve the gedankenexperiment, maybe. We place the mirror at the end of a long pole and extend it out 3 or 4 billion lightyears. Of course, we avoid gravity wells and we place it along an isotherm. This should give us the null relative velocity necessary. If galaxies are essentially unaffected by expansion, than I assume the pole's tensile strength is even less affected. However, if light is affected by the expansion by streeetching the wave, as some say, then it will be affected and return reddened; yet I say green.


You can get it to return green if you set up the problem that way, but you cannot get that result to emerge naturally using any mirrors that you will be likely to find without a starship attached to them. A starship(!?), no need to get elaborate with the props. ;)

Ken G
2007-Feb-12, 03:14 PM
We place the mirror at the end of a long pole and extend it out 3 or 4 billion lightyears out. Of course, we avoid gravity wells and we place it along an isotherm. This should give us the null relative velocity necessary. If galaxies are essentially unaffected by expansion, than I assume the pole's tensile strength is even less affected. However, if light is affected by the expansion by streeetching the wave, as some say, then it will be affected and return red; yet I say green.


Now I see what you're trying to do. That's a cute problem, and I suspect quite difficult for any but an expert. The problem is, the end of the rod will be moving quite relativistically wrt the comoving galaxies in which the cosmological principle may be applied, so is length contracted in a time-varying way in the simplest coordinates. There isn't a global inertial frame in the presence of all that universal gravity, so you can't say it's an SR problem and get no redshift. My guess is, there will be redshift, but much less than you'd get if the mirror was embedded in the distant galaxy. I think this would most naturally be viewed as a gravitational redshift, due to the time varying gravity in the universe as a whole.

George
2007-Feb-12, 05:42 PM
Yes, I failed to add the isograv constraint. :razz:

Of course, the purpose of this gedankenexperiment is to serve as a way to picture the distinction between a Doppler model and one which incorporates the more mainstream cause for redshift, the cosmological redshift. Admittedly, I am not that sharp on either one. [Maybe that explains why I relate to green better. :)]

Allowing for my magical isograv, would you guess green?

Jeff Root
2007-Feb-12, 05:46 PM
The problem is, the end of the rod will be moving quite relativistically
wrt the comoving galaxies in which the cosmological principle may be
applied, so is length contracted in a time-varying way in the simplest
coordinates.
Is that a problem? George isn't interested in the fact that other
galaxies may happen to be whizzing past the mirror at ludicrous
speeds. They aren't relevant to the question. All that is relevant
is the enormous green laser on my left, the long, long pole I'm holding,
the mirror at the end of the pole, and the spectrometer George is
looking through, on my right.

George, it seems like I've been holding this pole forever! Are you
almost done? My deltoid muscles are pooped!



There isn't a global inertial frame in the presence of all that universal
gravity, so you can't say it's an SR problem and get no redshift. My
guess is, there will be redshift, but much less than you'd get if the
mirror was embedded in the distant galaxy. I think this would most
naturally be viewed as a gravitational redshift, due to the time
varying gravity in the universe as a whole.
This sounds like something I read quite a few years ago, that only
something like one or two percent of the redshift is due to gravity,
and the remainder is due to Doppler. But you have disagreed with
that characterization when I stated it here previously.

-- Jeff, in Minneapolis

Ken G
2007-Feb-12, 05:51 PM
Of course, the purpose of this gedankenexperiment is to serve as a way to picture the distinction between a Doppler model and one which incorporates the more mainstream cause for redshift, the cosmological redshift. The key point in all this is that you cannot find an absolute "distinction" in interpretations, there is no distinction, there is only the calculation you do in a particular coordinate system and the interpretation you give it to be consistent with those coordinates. I do think this gedankenexperiment is a very important instructive example that I wager only a true GR expert can properly calculate, even assuming there is not some surprising problem associated with the "rigidity" of the rod you are imagining. Rigidity is really a nasty concept in relativity, for example it is impossible to rotate a perfectly rigid disk (because of the required length contraction).

I really try to avoid rigidity issues, and we may be bumping right into them here. But assuming that it is possible to have a rigid rod such as you imagine, and if you have a way to maintain a fixed gravitational potential all along the rod as the light propagates along, then I would indeed guess green in that situation. Note this would require some pretty significant local gravity sources to offset the expansion, and the end of the rod would really be slicing through whatever local galaxies were zooming by, but you seem to me to have controlled away all the elements that could lead to redshift. It's an interesting approach to the question, actually-- instead of asking "why is there redshift" you are asking "what would you need to not have redshift".

George
2007-Feb-12, 08:39 PM
George, it seems like I've been holding this pole forever! Are you almost done? My deltoid muscles are pooped! Yes, the laser beam is now on its way back and the mirror is no longer needed. Just in time, too, because the isothermal, isograv region is collapsing and will soon jerk your arm out of socket, or did I fail to mention this little quirk? :)


Rigidity is really a nasty concept in relativity, for example it is impossible to rotate a perfectly rigid disk (because of the required length contraction). I would assume the flex in the molecules would suffice, or does "perfectly rigid" disallow such. Or do you mean an enormous disk. I am unfamiliar with this idea but it is a sidebar to the gedankenexp.


It's an interesting approach to the question, actually-- instead of asking "why is there redshift" you are asking "what would you need to not have redshift". I wanted an understandable mental model that would distinguish Doppler redshift from the cosmological redshift. Assuming the expansion does stretch light, ie cosmological redshift, then the green light should return redshifted as it was subjected to the expansion effect during our multi-billion year wait (and its zero time wait). Though slightly impractical and difficult for Jeff, the pole, of course, locks the coordinate system down without even having to describe it.


...then I would indeed guess green in that situation. Hmmm, we agree, so I'll guess I'll put away the ice cream, then. :) Yet, is this not counter to mainstream and the idea that photons get stretched by expansion itself, or is it my misunderstanding of their photonic pedagogory?

Back at the pole, if this coordinate system could actually be plotted in deep space, then the mirror is no longer needed. If we observe doubly ionized oxygen from a distance of, say, z > 1 knowing at the time it had no relative motion to our current position (as if the pole were in place), then any redshift not attributable to gravity would be from the expansion itself. Of course, there is no chance of such an alignment, but if we know the difference in the speed of the green cloud emitter relative to our coordinate system, then we can us SR to adjust and see if it is green. I don't really know, though, so consider it to be in question form, or "is the above a possibility".

Ken G
2007-Feb-12, 09:36 PM
I wanted an understandable mental model that would distinguish Doppler redshift from the cosmological redshift. Assuming the expansion does stretch light, ie cosmological redshift, then the green light should return redshifted as it was subjected to the expansion effect during our multi-billion year wait (and its zero time wait). But what I'm trying to say is that the phrase "exposed to the expansion effect" is not a physically meaningful statement, contrary to what you may often hear. It's like if I asked you the question, when a light source is moving away from a receiver in special relativity, does the light redshift when it leaves the source, or when it arrives at the receiver, or gradually as due to being "exposed to the expansion effect". So you can't even give unique meaning to this phrase in special relativity, there is nothing in general relativity that suddenly does give it meaning. It's a coordinate interpretation, the physics is all just the redshift, for whatever reason.

I think when people talk about "the expansion effect" what they really mean is "hey we have this cosmological principle here, so we'd be simply nuts to choose any coordinatization other than the comoving frame, which has the additional advantage that all times are proper times, and in that particular frame the redshift interpretation is one of an "expansion effect". Either this takes too long to say each time, or people are simply not that nuanced. But note your puzzle breaks the cosmological principle in two places (ouch), once when it has the rod moving quite peculiarly relative to the galaxies, and again when you insert a mass distribution to maintain a fixed gravitational potential over billions of years. Maybe the latter isn't even possible for some deep reason involving what you can really do with gravity, I can't say much there it's too advanced.

Yet, is this not counter to mainstream and the idea that photons get stretched by expansion itself, or is it my misunderstanding of their photonic pedagogory?That's what I'm answering above-- no it's not counter to mainstream, but everyone should really spend a little more time identifying the coordinates they are using and learning that physical interpretations have always been coordinate dependent, even in Newtonian mechanics.

If we observe doubly ionized oxygen from a distance of, say, z > 1 knowing at the time it had no relative motion to our current position (as if the pole were in place)The simple presence of the pole is not the same thing as saying there's no relative motion, you need both the pole and what you are doing to gravity, the pole by itself does not have "no relative motion" because of length contraction effects in whatever coordinatization is being used. Relative motion is "relative" to a lot of things in GR, and there is no global inertial coordinatization in GR. Still, I think it is reasonable to break the redshift into two parts-- what you'd get for such a long pole (with no weird mass distributions to monkey with gravity), and what you actually get for comoving galaxies. If you do that, you might be tempted to associate the first with the gravitational effects, and everything else with the "relative motion", but it still wouldn't be a unique definition of that latter concept. It could very well be a useful one though, it's just a little beyond my knowledge of GR.

George
2007-Feb-12, 11:10 PM
It's a coordinate interpretation, the physics is all just the redshift, for whatever reason. I feared as much, but I curious about the component reasons for a given net redshift, nevertheless.


I think when people talk about "the expansion effect" what they really mean is "hey we have this cosmological principle here, so we'd be simply nuts to choose any coordinatization other than the comoving frame, which has the additional advantage that all times are proper times, and in that particular frame the redshift interpretation is one of an "expansion effect".Yes, a double dose of my fear. Either way produces the same net result, apparently. Of course, a closer look might reveal a discernable difference such as in the equality (or as my wife often says - "exactly the same but different"), but still different when causality is added, found in the question, "Does an apple fall from a tree or do we rise to the apple?".


But note your puzzle breaks the cosmological principle in two places (ouch), once when it has the rod moving quite peculiarly relative to the galaxies, and again when you insert a mass distribution to maintain a fixed gravitational potential over billions of years. Maybe the latter isn't even possible for some deep reason involving what you can really do with gravity, I can't say much there it's too advanced.Yes, no doubt you're right. Btw, does a good gedankenexperiment require no magic? I think it does not allow it. Anyway, mine does but only to help with the idea of superimposing a specific non-co-moving coordinate system over the co-moving one.

If Jeff will put the pole back up... let us now send two green pulses at the mirror. Since they are traveling in the co-moving frame they will return at a longer interval of time by our clock. But, I suppose, both pulses will still be green (if in the magical isograv and isotherm). This gets us closer to Kaptain K's comment but I think he, and others, will expect a redshifted color, maybe.


The simple presence of the pole is not the same thing as saying there's no relative motion, you need both the pole and what you are doing to gravity, the pole by itself does not have "no relative motion" because of length contraction effects in whatever coordinatization is being used. Agreed, I confess to cheating gravity its influence. :)


Still, I think it is reasonable to break the redshift into two parts-- what you'd get for such a long pole (with no weird mass distributions to monkey with gravity), and what you actually get for comoving galaxies. If you do that, you might be tempted to associate the first with the gravitational effects, and everything else with the "relative motion", but it still wouldn't be a unique definition of that latter concept. It could very well be a useful one though, it's just a little beyond my knowledge of GR.It is way over my level. Since the co-moving coordinate system has gravity decreasing then I can understand a gravitational redshift component. Yet, I tend to think others are adding another component attributable more to the co-moving coordinate system itself.

Ken G
2007-Feb-13, 06:38 AM
Btw, does a good gedankenexperiment require no magic? I think it does not allow it. Anyway, mine does but only to help with the idea of superimposing a specific non-co-moving coordinate system over the co-moving one.Yes, a true gedankenexperiment must be limited to what is possible in principle, which is different from magic. So whether yours is a good gedanken or not depends crucially on whether or not it is really possible, in principle, to maintain a fixed gravitational potential along the rod. I just don't know, but if it is possible, then I do think it's a reasonable way to conceptually isolate the "most natural" gravitational redshift part, with all the remaining redshift in a real situation being due to Doppler shifts. It's all pedagogical of course, but a good pedagogy is a good thing.


If Jeff will put the pole back up... let us now send two green pulses at the mirror. Since they are traveling in the co-moving frame they will return at a longer interval of time by our clock.No, I would say that the action of maintaining a fixed gravitational potential will mean that the time interval will not be longer, and that's why they'll still be green. There is no "absolute" time interval, it depends on what gravity is doing.

Since the co-moving coordinate system has gravity decreasing then I can understand a gravitational redshift component. Yet, I tend to think others are adding another component attributable more to the co-moving coordinate system itself.
I think many people attribute more "physical meaning" to "expanding space" than is really there. It's just a particular way of interpreting what is going on, it's not a physical rule that light stretches with time, just as it wasn't in the purely Doppler example I mentioned. However, it is a particularly useful interpretation, and is most commonly seen, it just gets taken too seriously.

George
2007-Feb-13, 03:06 PM
Yes, a true gedankenexperiment must be limited to what is possible in principle, which is different from magic. Ok. [I assume there is no cool name for one that oversimplifies to the point of requiring some magic.]


So whether yours is a good gedanken or not depends crucially on whether or not it is really possible, in principle, to maintain a fixed gravitational potential along the rod. I just don't know, but if it is possible, then I do think it's a reasonable way to conceptually isolate the "most natural" gravitational redshift part, with all the remaining redshift in a real situation being due to Doppler shifts. It's all pedagogical of course, but a good pedagogy is a good thing. I suppose a zillion rockets could accelerate accordingly but how would anyone know just when and where to place them in warped space is far beyond me. Of course, they would have plenty of time to figure it out before the lengthy pole was in place. :)


No, I would say that the action of maintaining a fixed gravitational potential will mean that the time interval will not be longer, and that's why they'll still be green. I understand that gravity only slows the expansion and not responsible for it. As in the balloon analogy, as space expands it must increase the distance between the pulses because space has oozed in between them. I am pretty "strawberry sundae sure" about this, too. :)


I think many people attribute more "physical meaning" to "expanding space" than is really there. It's just a particular way of interpreting what is going on, it's not a physical rule that light stretches with time, just as it wasn't in the purely Doppler example I mentioned. However, it is a particularly useful interpretation, and is most commonly seen, it just gets taken too seriously.I can not picture the alternative to an expanding space that doesn't either stretch or ooze forth into space as we know it, though that may due to oversimplifying the idea in those general readership books I have read. I won't offer to bet as I can not demonstrate adequately that my understanding is correct, though I might go for a majority vote as an alternative; hardly fair but I do like sundaes. :)

Ken G
2007-Feb-13, 03:20 PM
Ok. [I assume there is no cool name for one that oversimplifies to the point of requiring some magic.]
Sure there is-- "science fiction". :)

I understand that gravity only slows the expansion and not responsible for it. As in the balloon analogy, as space expands it must increase the distance between the pulses because space has oozed in between them.Indeed, I think the easiest way to think about cosmology is to get rid of gravity altogether. It plays no essential role for understanding the Big Bang, it is only needed for quantitative work. And your statement about space "oozing in" forcing the distance to increase still seems to me like you are reversing the logic-- the space oozes in because the distance is increasing.

I can not picture the alternative to an expanding space that doesn't either stretch or ooze forth into space as we know it, though that may due to oversimplifying the idea in those general readership books I have read. Here's one simple alternative-- just let matter and bound systems shrink in place. The "oozing and stretching" is now quite different-- interpretations are not unique.

Kaptain K
2007-Feb-13, 04:46 PM
Sure there is-- "science fiction".
Make that "science fantasy".

Ken G
2007-Feb-13, 05:28 PM
Good point-- "the force" changes Star Wars from fiction to fantasy.

George
2007-Feb-13, 09:36 PM
And your statement about space "oozing in" forcing the distance to increase still seems to me like you are reversing the logic-- the space oozes in because the distance is increasing. That sounds like vacuum power, which should make sense as my logic system and all the dozens of synapsis are in one. :) However, I abhor a vacuum (so does someone else, IIRC); partly because I am so intimidated by virtually everything that's in it. ;)


Here's one simple alternative-- just let matter and bound systems shrink in place. The "oozing and stretching" is now quite different-- interpretations are not unique. :surprised [Speaking of science fiction. :)] I'm not that gullible with Gulliver. I perfer the apple to fall from the tree.



Make that "science fantasy".Good point-- "the force" changes Star Wars from fiction to fantasy. :rolleyes: and we're not even close to Friday. :) How 'bout, "genobbinexperiment"; hey, it sounds cool, right?