Is the light from Andromeda truly blue shifted?

[sol47739]

TL;DR Summary

Is the light from Andromeda truly blueshifted or just less redshifted than the most other objects?

I asked a similar question on Quora not that long ago:https://www.quora.com/Is-the-light-...e&__sncid__=37627098260&__snid3__=50480233916
I got an answer saying that the light isn’t blue shifted but redshifted but less redshifted than most other objects. I think this sounds reasonable considering the vast distances. Do you know of any papers, experiments, data confirming this?

 

[fresh_42]

How about https://ui.adsabs.harvard.edu/abs/2015MNRAS.451..724W/abstract ?

 

[sol47739]

How about https://ui.adsabs.harvard.edu/abs/2015MNRAS.451..724W/abstract ?

Thanks! I will look at it.

 

[Ibix]

Andromeda is coming towards us, and is hardly any distance away in cosmological terms. Even the Wikipedia page notes that its redshift is negative.

 

[fresh_42]

Here is another one:

https://ui.adsabs.harvard.edu/abs/1982MNRAS.199...67E/abstract

 

[sol47739]

Andromeda is coming towards us, and is hardly any distance away in cosmological terms. Even the Wikipedia page notes that its redshift is negative.

Yes negative in comparison to other cosmological objects but not „actually“ negative.

 

[phyzguy]

Yes negative in comparison to other cosmological objects but not „actually“ negative.

No! Its redshift is "actually" negative. It's redshift of -0.001 means it is moving toward us at 0.001*c ~=300 km/sec. The Hubble constant is about 70 km/sec/Mpc. At Andromeda's distance of 0.75 Mpc, the cosmological redshift is about 50 km/sec, which is much smaller than the 300 km/sec that it is moving toward us. Also, because Andromeda is gravitationally bound into the local group, the Hubble recession doesn't really apply.

 

[Ibix]

Yes negative in comparison to other cosmological objects but not „actually“ negative.

No, just negative. $z=0$ means stationary with respect to us; positive means receding and negative means approaching.

 

[sol47739]

No! Its redshift is "actually" negative. It's redshift of -0.001 means it is moving toward us at 0.001*c ~=300 km/sec. The Hubble constant is about 70 km/sec/Mpc. At Andromeda's distance of 0.75 Mpc, the cosmological redshift is about 50 km/sec, which is much smaller than the 300 km/sec that it is moving toward us. Also, because Andromeda is gravitationally bound into the local group, the Hubble recession doesn't really apply.

So you can see on the emission lines that they are higher in frequency than the usual emission lines? Or is it that almost everything in space is redshifted to lower frequency and the Andromeda is just less redshifted than predicted by the cosmological redshift?

 

[sol47739]

So you can see on the emission lines that they are higher in frequency than the usual emission lines? Or is it that almost everything in space is redshifted to lower frequency and the Andromeda is just less redshifted than predicted by the cosmological redshift?

Isn’t the answer I got from the person on Quora correct? https://www.quora.com/Is-the-light-...e&__sncid__=37627098260&__snid3__=50480233916

 

[sol47739]

No! Its redshift is "actually" negative. It's redshift of -0.001 means it is moving toward us at 0.001*c ~=300 km/sec. The Hubble constant is about 70 km/sec/Mpc. At Andromeda's distance of 0.75 Mpc, the cosmological redshift is about 50 km/sec, which is much smaller than the 300 km/sec that it is moving toward us. Also, because Andromeda is gravitationally bound into the local group, the Hubble recession doesn't really apply.

From Quora:
The light from Andromeda galaxy is actually redshifted, not blueshifted. The light from all distant galaxies appears to be redshifted, meaning that the light waves are stretched out and shifted towards the red end of the spectrum. This redshift is caused by the expansion of the universe, which causes the galaxies to move away from each other, causing the light waves to stretch out.

However, it is possible for the redshift to be less than expected, which would make the light appear blueshifted in comparison to the rest of the redshifted light from other distant galaxies. This can happen if Andromeda is moving towards us faster than expected or if there are other factors causing a blueshift in the light, such as gravitational lensing.

In any case, the light from Andromeda is still predominantly redshifted and not blueshifted.

Isn’t this correct?

 

[Ibix]

So you can see on the emission lines that they are higher in frequency than the usual emission lines?

Yes.

Or is it that almost everything in space is redshifted to lower frequency and the Andromeda is just less redshifted than predicted by the cosmological redshift?

Andromeda is a member of the Local Group of galaxies, which is gravitationally bound and not expanding.

Isn’t the answer I got from the person on Quora correct?

No, it is not correct. Andromeda is not a distant galaxy - it's the nearest one that isn't a satellite of the Milky Way.

 

[Hyperfine]

So you can see on the emission lines that they are higher in frequency than the usual emission lines? Or is it that almost everything in space is redshifted to lower frequency and the Andromeda is just less redshifted than predicted by the cosmological redshift?

As stated by others, M31 is blue shifted: z = -0.001004; velocity -301 km/s.

See: NASA/IPAC Extragalactic Database entry Messier 031

 

[malawi_glenn]

From Quora:

Why are you using Quora as a reference? You have been given plenty of scientific references in this thread.

 

[Ibix]

Isn’t the answer I got from the person on Quora correct? https://www.quora.com/Is-the-light-...e&__sncid__=37627098260&__snid3__=50480233916

Looking at this again, I think the Quora poster is making a mess of explaining something that's true. Galaxies move away from us faster the further away they are. But the universe is not exactly a naive Friedman-Lemaitre-Robertson-Walker (FLRW) model where everything exactly obeys a perfect Hubble law. Everything has a random velocity compared to what it would have in a perfect FLRW model, so they have a small extra red or blue shift. But the average magnitude of those velocities doesn't really change with distance and the recession speed only grows with distance - so the random contribution can easily exceed the modelled cosmological redshift at short distances, while being only a tiny error term on very distant objects. (In fact, nearby objects are often gravitationally bound to one another and cosmological expansion doesn't apply within the system at all).

We're doing about 600km/s compared to a hypothetical perfect FLRW galaxy. Back of the envelope, then, recession speed is comparable to this at around ten megaparsecs, so that's the kind of distance you'd expect to see everything redshifted - nearer stuff will often be blue shifted. Andromeda is less than one megaparsec away and it happens to be coming our way, so is blue shifted.

 

[PeroK]

Or, you could look at the data. And see what spectroscopy tells you!

 

[Hyperfine]

Looking at this again, I think the Quora poster is making a mess of explaining something that's true. Galaxies move away from us faster the further away they are. But the universe is not exactly a naive Friedman-Lemaitre-Robertson-Walker (FLRW) model where everything exactly obeys a perfect Hubble law. Everything has a random velocity compared to what it would have in a perfect FLRW model, so they have a small extra red or blue shift. But the average magnitude of those velocities doesn't really change with distance and the recession speed only grows with distance - so the random contribution can easily exceed the modelled cosmological redshift at short distances, while being only a tiny error term on very distant objects. (In fact, nearby objects are often gravitationally bound to one another and cosmological expansion doesn't apply within the system at all).

We're doing about 600km/s compared to a hypothetical perfect FLRW galaxy. Back of the envelope, then, recession speed is comparable to this at around ten megaparsecs, so that's the kind of distance you'd expect to see everything redshifted - nearer stuff will often be blue shifted. Andromeda is less than one megaparsec away and it happens to be coming our way, so is blue shifted.

All well and good. However, the shift is a direct spectroscopic observable. The data indicate unambiguously that in the case of M31 the shift is to the blue.

 

[TeethWhitener]

The OP might use Hubble's law to calculate the expected recession velocity for Andromeda, and compare it to the measured recession velocity.

 

[phinds]

The OP might use Hubble's law to calculate the expected recession velocity for Andromeda, and compare it to the measured recession velocity.

??? What recession velocity? Andromeda is not receding from us, it is approaching us. Also, Hubble's Law does not apply to gravitational bound objects.

 

[Ibix]

All well and good. However, the shift is a direct spectroscopic observable. The data indicate unambiguously that in the case of M31 the shift is to the blue.

Oh, absolutely. The number that comes out of the spectrometer is unavoidable. You can break it down as a "cosmological redshift" and a "peculiar velocity red/blue shift" if you want (you can do that for your toes if you want, even if it makes no real sense to do so) and say some object is less/more redshifted than it "should" be, but the number is what it is and in this case it's blue shifted.

 

[TeethWhitener]

??? What recession velocity? Andromeda is not receding from us, it is approaching us. Also, Hubble's Law does not apply to gravitational bound objects.

Maybe you should do the same calculation to see my point.

 

[phinds]

Maybe you should do the same calculation to see my point.

What is the point of trying to apply Hubble's Law in a domain where it does not apply?

 

[TeethWhitener]

What is the point of trying to apply Hubble's Law in a domain where it does not apply?

Alright fine, I guess I'll do the homework for you.

The point is that, even if you took Hubble's law as gospel, it predicts a recessional velocity of around +50 km/s for Andromeda. The measured velocity relative to the Milky Way is around -300 km/s toward us. The point I was trying to make is that it doesn't matter whether the comoving velocity is included or not. You still get a negative number at the end of the day.

The broader point is that of course Hubble's law applies (just like all the rest of GR). It's just that gravity is a much much much larger effect than spacetime expansion at distances less than 100 Mpc or so.

 

[Hyperfine]

TL;DR Summary: Is the light from Andromeda truly blueshifted or just less redshifted than the most other objects?

The question posed is quoted above. The answer is given by the data, not by some interpretation of why the data is what it is.

 

[TeethWhitener]

The broader point is that of course Hubble's law applies (just like all the rest of GR). It's just that gravity is a much much much larger effect than spacetime expansion at distances less than 100 Mpc or so.

It occurs to me just now that I don't actually know this. GR being so nonlinear (and me being so non-expert), I don't actually know at what point a Schwarzschild perturbation on a FLRW background becomes non-perturbative. So I can't really say the above with any confidence.

 

[Hyperfine]

The point is that, even if you took Hubble's law as gospel, it predicts a recessional velocity of around +50 km/s for Andromeda. The measured velocity relative to the Milky Way is around -300 km/s toward us. The point I was trying to make is that it doesn't matter whether the comoving velocity is included or not. You still get a negative number at the end of the day.

I find this argument to be circular. The measured velocity is determined from the observed spectral shifts.

Perhaps this is nothing more than a semantic point, depending on what one means when referring to redshift, or blueshift. Does it refer explicitly to the spectroscopic observations, or does it denote motion relative to the observer? Personally I would say the spectroscopic observations, taking the concept of relative motion to be an interpretation of the observational data.

 

[Ibix]

I find this argument to be circular.

You can talk about the redshift one would see from a hypothetical comoving source colocated with Andromeda, and whether the light you receive from Andromeda is more or less redshifted than that.

I suspect what @TeethWhitener is getting at is that the velocities of galaxies relative to colocated comoving observers are typically a few hundred kilometres per second, so you may well see blueshifted sources within a radius of a few hundred kilometres per second divided by the Hubble constant. Beyond that radius you'll see fewer and fewer blueshifted galaxies.

 

[TeethWhitener]

I find this argument to be circular. The measured velocity is determined from the observed spectral shifts.

Perhaps this is nothing more than a semantic point, depending on what one means when referring to redshift, or blueshift. Does it refer explicitly to the spectroscopic observations, or does it denote motion relative to the observer? Personally I would say the spectroscopic observations, taking the concept of relative motion to be an interpretation of the observational data.

My point was this: say you looked up the blueshift of Andromeda from NASA or wherever (I'm using velocity as a proxy: -300 km/s) and you had no idea whether NASA had a convention to take the spectral shift as absolute or relative to what is expected for that recessional velocity at that distance. If you work out the expected recessional velocity from Hubble's law, you find that the NASA number must indicate an actual blueshift, regardless of whether the reported number is relative or absolute, because the recessional velocity is only +50km/s (so Andromeda's velocity is either -300 km/s or -250 km/s, but in either case it is negative and its light is therefore blueshifted). I'm sorry if that point wasn't clear.

 

[Hyperfine]

My point was this: say you looked up the blueshift of Andromeda from NASA or wherever (I'm using velocity as a proxy: -300 km/s) and you had no idea whether NASA had a convention to take the spectral shift as absolute or relative to what is expected for that recessional velocity at that distance. If you work out the expected recessional velocity from Hubble's law, you find that the NASA number must indicate an actual blueshift, regardless of whether the reported number is relative or absolute, because the recessional velocity is only +50km/s (so Andromeda's velocity is either -300 km/s or -250 km/s, but in either case it is negative and its light is therefore blueshifted). I'm sorry if that point wasn't clear.

Thanks for the clarification.

 

[PeroK]

It occurs to me just now that I don't actually know this. GR being so nonlinear (and me being so non-expert), I don't actually know at what point a Schwarzschild perturbation on a FLRW background becomes non-perturbative. So I can't really say the above with any confidence.

I don't think it's valid to apply Hubble's law at all, because the local group of galaxies is not a region of space with the average energy density of the universe as a whole. Hubble's law applies to the universe at the largest scale, where the average energy density applies, but not to every region of the universe, where different equations govern the local galactic kinematics.

Interestingly, Hubble derived his law using data for relatively close galaxies. How did he do this? Here's what Steven Weinberg says in The First Three Minutes:

His conclusion was that there is a "roughly linear relation" between velocities and distances. Actually, a look at Hubble's data leaves me perplexed how he could reach such a conclusion - galactic velocities seem almost uncorrelated with their distance ... In fact, we would not expect any neat relation of proportionality between velocity and distance for these 18 galaxies - they are all much too close. It is difficult to avoid the conclusion that ... Hubble knew the answer he wanted to get.

 

[TeethWhitener]

I don't think it's valid to apply Hubble's law at all, because the local group of galaxies is not a region of space with the average energy density of the universe as a whole. Hubble's law applies to the universe at the largest scale, where the average energy density applies, but not to every region of the universe, where different equations govern the local galactic kinematics.

Interestingly, Hubble derived his law using data for relatively close galaxies. How did he do this? Here's what Steven Weinberg says in The First Three Minutes:

His conclusion was that there is a "roughly linear relation" between velocities and distances. Actually, a look at Hubble's data leaves me perplexed how he could reach such a conclusion - galactic velocities seem almost uncorrelated with their distance ... In fact, we would not expect any neat relation of proportionality between velocity and distance for these 18 galaxies - they are all much too close. It is difficult to avoid the conclusion that ... Hubble knew the answer he wanted to get.

The average density of the Local group is about twice the average density of the universe (Ref1 Ref2). I’m curious now as to whether that counts as a big perturbation to the FLRW metric. My gut says probably but I’m not an expert.

Which is what I was unsure about: when you say Hubble’s law doesn’t apply, do you mean that the recessional velocity is insignificant/of the same order as local astronomical velocities, or do you mean that, because GR is nonlinear, the metric of an FLRW background plus local matter is sufficiently qualitatively different from the flat FLRW metric that Hubble’s law, or an analog of it, can no longer be derived from the metric?

 

[PeroK]

Which is what I was unsure about: when you say Hubble’s law doesn’t apply, do you mean that the recessional velocity is insignificant/of the same order as local astronomical velocities, or do you mean that, because GR is nonlinear, the metric of an FLRW background plus local matter is sufficiently qualitatively different from the flat FLRW metric that Hubble’s law, or an analog of it, can no longer be derived from the metric?

That's a good question. Hubble's law requires an evolution of the universe from the earliest time. In order for Andromeda to be moving away from the Milky Way at recession speed $v$ requires a long time evolution. It's not like a gravitational force that acts directly.

Look at it this way. Take two objects (Earth and Sun, say) and look at their behaviour today. Then, imagine their behaviour at a time in the future when the universal expansion is enormous. It makes no difference. The local behaviour will not change over time as the universe globally expands more rapidly. Unless the system remains part of the Hubble flow for billions of years, it simply does not have the accelerating background average recessional velocity. Not at all.

 

[Bandersnatch]

Interestingly, Hubble derived his law using data for relatively close galaxies. How did he do this? Here's what Steven Weinberg says in The First Three Minutes:

I'm a bit late to the party, but let me address this.
While today any small-scale region is most certainly not uniform, there was a time when it was. Galaxies that we see today are made of material that must have necessarily inherited the initial Hubble flow. It's not like all that matter hit brakes the moment the deviation from uniform distribution reached some arbitrary threshold.

Over the history of the universe, many objects have managed to bleed off the initial impulse and coalesce, collide, combine, begin approaching or orbiting one another. But the farther apart and smaller the objects, the longer it takes. For most of the galaxies used by Hubble in his paper, the dominant component of their motion is still the initial recession. Even though they are all technically in a single bound system.

And it shows when one plots them - the trend is clear, whether one uses Hubble's original, or modern, distance and velocity data (although it is more messy in the former case).
I'm not sure why Weinberg would say what he did there.

 

[mitchell porter]

I think the Quora poster is making a mess of explaining something that's true.

The Quora poster's answers are generated by ChatGPT.