PDA

View Full Version : Redshifts



Matthias
2007-Mar-03, 03:24 AM
How do you tell the difference between the redshift in light emitted by a star or galaxy receding from the earth at a high velocity, and the redshift in light caused by the light energy travelling through interstellar or intergalactic dust in between a star or galaxy and the earth?

Is it possible that the redshifts seen from the most distant of galaxies is a consequence of light being filtered through the intergalactic medium?

Matthias
2007-Mar-03, 03:47 AM
On second thought there looks like there's a whole bunch of threads about redshifting around here. Apologies if this thread was redundant.

EvilEye
2007-Mar-03, 03:50 AM
And...what about reflected light? Let's assume that a beam of light left one galaxy or star, and bounced off off some reflective property of another galaxy or planet....

Would we interpret it as where it came from? Or could we know that it was reflected?

Matthias
2007-Mar-03, 03:55 AM
Sometimes I wonder if we would ever be able to know if we observed a galaxy whose image was gravitationally distorted by two other galaxies before its light got here.

Ken G
2007-Mar-03, 04:19 AM
Redshift by dust has a very different signature than relativistic redshift, they are easy to distinguish.

Amber Robot
2007-Mar-03, 05:09 AM
How do you tell the difference between the redshift in light emitted by a star or galaxy receding from the earth at a high velocity, and the redshift in light caused by the light energy travelling through interstellar or intergalactic dust in between a star or galaxy and the earth?

Interstellar dust doesn't cause redshift. The light might get "reddened", but that means that the blue light is extincted more than the red light, but the wavelengths of spectral features are essentially unchanged by the extinction.

trinitree88
2007-Mar-05, 02:33 AM
How do you tell the difference between the redshift in light emitted by a star or galaxy receding from the earth at a high velocity, and the redshift in light caused by the light energy travelling through interstellar or intergalactic dust in between a star or galaxy and the earth?

Is it possible that the redshifts seen from the most distant of galaxies is a consequence of light being filtered through the intergalactic medium?

Matthias. Actually the two redshifts that are indistinguishable are Doppler-Fizeau and gravitational. One needs to know the mass of the emitting object vs the mass of the object on which your frequency detector is mounted (like the Earth) and the radial velocity between the two. It's actually quite tricky.
In addition conservation of momentum, which applies to photons as well as masses, forbids a photon from losing energy and momentum unless there is something to lose it to. In the case of the photon in intergalactic space, it's not "empty"; it could lose both in a neutral current interaction with the neutrino sea since it could only see neutrinos from the forward light cone. Pete.

Nereid
2007-Mar-06, 06:52 PM
trinitree88 is right, for a point source.

For an extended source, it is quite easy to estimate the contribution from a gravity well ... as long as you assume GR applies (anyone like to have a shot at explaining why, and how?)

The two sources of redshift that are hardest to tease apart are that due to the expansion of the universe and that due to relative, 'local', motion. Anyone like to have a shot at explaining why it's so hard to tease the two apart, and how you might go about doing it?

speedfreek
2007-Mar-07, 01:00 AM
The way I understand it is thus:

Redshift is an expression of the shifting of absorption and emission lines in the measured spectrum of an object. To measure redshift (where redshift is expressed as the quantity z) we use

z = (observed wavelength - emitted wavelength) / emitted wavelength
or
1 + z = observed wavelength / emitted wavelength

If z<0 then the spectrum is blue-shifted, if z>0 then it is redshifted.

As has been said above, light-matter interactions such as with dust or scattering result in energy shifts in the radiation field and are generally referred to as "reddening". This means the whole spectrum is moved, rather than the absorption and emission lines shifting within the spectrum, and thus is easy to detect.

Once a redshift measurement is made, it then has to be interpreted. This is done using different forms of mathematical transformation, depending on the type of object you are examining. This is the key. Basically, we have to decide what type of redshift(s) an object might have, and then apply the transformations to give us the relevent answers. Redshift alone cannot tell us the story, we have to use our current assumptions about an object to work out the values. If there are 2 mechanisms that may impart redshift on an objects spectrum we have to use other means to work out what proportion is caused by what form of redshift.

If z > 0.1 then the redshift is almost totally dominated by cosmological redshift, caused by the expansion of space. Galaxies aren't moving very much, inertially, relative the apparent movement caused by the expansion of space. Galaxies seem to have a tendency to cluster together and move around each other a bit, but this movement is very small when compared to the amount they seem to have moved away from us. There, I said it twice!

We know of a few different forms of redshift: Doppler redshift (also known as Doppler-Fizeau effect), Relativistic Doppler Redshift (which is a more complete form of doppler, taking relativity into account), Cosmological redshift and Gravitational redshift.

Gravitational red shifts are generally very small, and you only get very large ones from the light emitted near neutron stars or black holes - environments you can independently confirm from other observations. With the exception of the sun, no gravitational red shifts have been detected for ordinary stars, but they ought to be present if we had good enough instruments.

Mainly, to distinguish gravitational redshifts from other kinds, you compare the size of the object with its mass to determine how much larger it is than its black hole radius. Objects like nebulae and entire galaxies are trillions of times larger than their BH radius, so the magnitude of the redshift is 1 part in a trillion. Normal stars are only a few hundred thousand times larger than their BH radius, so light from their surfaces is at the limit of being able to detect, spectroscopically, such a gravitational redshift. Neutron stars and white dwarfs are about 10, and 3000 times larger than their BH size so gravitational redshifts are of the order of 1 part in 10 to 1 part in 1000.

Cosmological redshifts are only important and easily distinguishable for rather distant galaxies, but can get mixed up with the Doppler shift from the regular spatial motions of galaxies. Cosmological redshifts are only seen unambiguously at distances of billions of light years. At nearer distances, ordinary Doppler shifts from galaxian motion with respect to a local center of mass (galaxy cluster) is comparable to the cosmological effect and you have to disentangle the two contributions very carefully.

Ken G
2007-Mar-07, 10:45 AM
The two sources of redshift that are hardest to tease apart are that due to the expansion of the universe and that due to relative, 'local', motion. Anyone like to have a shot at explaining why it's so hard to tease the two apart, and how you might go about doing it?
It's not possible in general, but you can do it once you have defined what you mean by the redshift due to expansion. If one takes the latter to mean the comoving-frame redshift, then the former emerges as the variance on top of that average redshift. So you'd look for material in the same physical vicinity, and associate any random or differential redshifts to local motions (once you've dispensed with intrinsic gravitational redshifts in the manner described by speedfreak). I feel it is important to stress, as I often do, that this is not a physically unique approach, it's just the way to be true to a particularly sensible choice of coordinates that moves smoothly with the average location of large groups of galaxies.

Nereid
2007-Mar-07, 02:05 PM
Pick any distance, as long as it's greater than 1 pc (say).

Pick a minimum angular resolution, over which the best you can do is an integrated spectrum*

Pick a minimum detectable gravitational redshift, expressed as km/sec perhaps, say 10 km/sec.

Plug in the numbers and I think you'll find that only an SMBH within a few (dozen?) pc would leave a gravitational redshift footprint in the 'extended source' matter in its environment. Needless to say, we'd have noticed any such nearby SMBH a long time ago, via its enormous influence on the motions of nearby stars (for example).

Within the next decade or so, it may be possible to observe the nearest SMBH (SgrA*) with sufficient resolution to directly detect a gradient of (gravitational) redshift across any accretion disk (or similar) around it.

And it's possible today to interpret line profiles of galactic nuclei as having a gradient of gravitational redshift component, but the nuclei are not resolved.

Finally, as speedfreek noted, gravitational redshift can be detected within our solar system, for an extended source ... the Sun.

Now if the Sun had a white dwarf binary companion ...

*This is trickier than it seems, but for most objects with good (spectral) lines, it's >0.01", and mostly >0.1" ... though I've not checked the radio spectrum, and VLBI.

Cougar
2007-Mar-07, 03:59 PM
How do you tell the difference between the redshift in light emitted by a star or galaxy receding from the earth at a high velocity, and the redshift in light caused by the light energy travelling through interstellar or intergalactic dust in between a star or galaxy and the earth?
As Amber Robot mentioned, the latter doesn't really cause redshift. It may cause reddening and dimming, but even this can be identified and quantified from the differential absorption of different colors of light.

Ken G
2007-Mar-07, 10:30 PM
Plug in the numbers and I think you'll find that only an SMBH within a few (dozen?) pc would leave a gravitational redshift footprint in the 'extended source' matter in its environment.

That's using the conventional definition of "gravitational redshift" as an anomalous redshift due to local variations in the mass distribution, analogous to "Doppler redshifts" due to local variations in velocity. It all presupposes that cosmological redshift is due to expansion of space, which happens in the comoving coordinate system. But if the question is "how do you know it isn't a gravitational redshift", it begs that question a little to presume comoving coordinates. I believe there are other coordinatizations where the cosmological redshift is itself a gravitational redshift, indeed that is probably what you get if you apply the same coordinates that people use when they say "time slows down near a black hole". When people say that, usually they are applying a kind of hybrid coordinatization, where the coordinates comove with freefalling matter on large scales but not on small scales, so you get "cosmological redshift" on large scales and "gravitational redshift" on small scales. This distinction is telling us more about the conventional choice of coordinates than it is saying anything physical, as I understand these issues. I don't dispute that yours is the conventional lexicon, I'm just pointing out that we are basically looking at ourselves in the mirror when we say these things-- we are looking at the way we like to coordinatize.

Nereid
2007-Mar-07, 10:42 PM
That's using the conventional definition of "gravitational redshift" as an anomalous redshift due to local variations in the mass distribution, analogous to "Doppler redshifts" due to local variations in velocity.

[snip]Yes ... though I'd rather make the analogy weak (and explore the conventional difference between 'local variations in velocity' and 'cosmological redshift' in another post).

My intent was simply to note that, as a practical matter, astronomical observations, today (with today's instruments and techniques) do not, and (based on what we understand to be locations of SMBHs) could not, observe 'gravitational redshift' in an extended source (well beyond the solar system).

Nereid
2007-Mar-09, 04:13 AM
Redshift is redshift is redshift ... the observed wavelength of H alpha (or LyA), from an astronomical source, is whatever it is, and the rest is interpretation ... or, if you prefer, the application of your favourite set of theories of physics.

Let's assume the astronomical object in question is an extended source, so, in principle at least, we could get the redshift at each resolved pixel.

We know that our observatory is moving, wrt the centre of mass of the Earth, and we have a pretty good idea of the size and direction of that motion, so we can adjust our observed redshift for it.

Ditto wrt the solar system barycentre.

So far, no serious issues about interpretation, right?

We have a (very good) theory about the universe, cosmological expansion, etc, etc, etc. We now have a value for H0 that seems OK, and an estimate of the uncertainty (systematic and random error) of that value.

That very good theory of the universe also allows us to interpret the CMB dipole, as motion of the solar system barycentre wrt the CMB (one stage in the data pipeline of CMB observations is 'adjusting for' motion of the detector wrt the solar system barycentre).

If we have a good estimate of the distance to our astronomical object, then estimating the contribution of the cosmological redshift to the total observed redshift is a piece of cake, right?


... and that's where it gets tricky: there aren't many methods that will yield a few% reliable estimate of distance, for single extragalactic (or extra-Local Group) objects.

And if we want to be particularly cautious, and insist on building in the full extent of uncertainty, all the way along the chain that leads to an estimate of the cosmological redshift component of an object beyond the Local Group, then the uncertainty can be quite large, particularly for objects as close as the Virgo cluster (for example), or even any object clearly in the grip of the Great Attractor.

Of course, taking the big picture - averaging over hundreds or millions of objects, for example - the cosmological redshift signal can be seen with great clarity.