Redshift

[higginsdj]

Hi all,

I imaged a galaxy cluster in BVRI then measuered their colour indicies (B-V, R-I etc). I plotted these indicies against distance (from the PGC catalogue) and found that the galaxies shifted to the blue end of the spectrum the further away they were. (B-V increased as did R-I but V-R remained basically unchanged)

Now Redshift shows that the further an object is away, the more it's spectral lines will shift to the red. Does this not also mean that the light should also shift to the red - ie B-V and R-I should decrease? Am I missing something here? The target fields were located at a distance of around 150mLy and 650mLy.

Cheers

David

[Hitchhiker]

I'm nowhere near qualified to answer this question, but could it be that the galaxy cluster you have imaged just happens to be moving towards us rather than away? Even though the Universe as a whole is expanding, on a local level, galaxies are moving in all directions. For example, the Andromeda galaxy is moving towards us due to local gravitational conditions and would show a blue shift rather than red.

BTW what galaxy cluster were you imaging?

[xelasnave]

Maybe you observed the very moment the Universe has started to colapse:rofl:
That is very interesting I await others input and will think upon it myself
alex

[vespine]

From what I understand, redshift is not something you can easily observe "visually"..

Most astronomical sources like stars emit LARGE bands of electromagnetic radiation, far wider then just the narrow "visible band" of the whole spectrum.

When something is red shifted, the WHOLE emission is shifted to the RED side of the spectrum.

As an example, that means visible RED emissions can red shift into the INVISIBLE infra red (this is actually AWAY from red, but it is towards the red END of the spectrum, as opposed to the blue end).
On the other end, the violet moves into the blue AND the previously invisible ULTRAVIOLET part of the emission moves INTO the visible part of violet.

All that has happened is that the WHOLE spectrum has shifted to the right, so the NET effect on the "WIDTH" of visible light is actual ZERO.

Now if that was the case, then HOW the hell could you know that red shift has even occurred?

Well, it is because the EMISSION BANDS of most atoms are very well known and it is those bands that give away the clue. If you can imagine that the --- below is a full visible spectrum of hydrogen and the full stops are little gaps in the spectrum that are visible with a spectrometer. A normal hydrogen's emission on earth looks like this (for example only):

BLUE----.-----.-.-.---------..-----RED
Then a RED shifted emission line will look like this:
BLUE-------.-----.-.-.---------..--RED
But visibly, they may not look very different at all...

[xelasnave]

However if there is an observed change relevant to the blue visual light that it curious even if it is not the same area as spectrum analysis.
alexx

[vespine]

Well, I don't mean to be a stick in the mud but even that isn't a given, what he has perceived is a difference in measured colour indices of an image. Unless someone has experienced this exact thing themselves I think it would be very hard to interpret anything from that by itself, too many unknowns in my opinion.

What was the sample size? Could you do the same in a different region of sky? Could someone with different equipment reproduce the result? Could be the CCD, or maybe it was just a coincidence? Have you tried drawing your graph to scale? Do the values line up?

[xelasnave]

Point taken
alex

[higginsdj]

The sample size was 20 of the 34 galaxies in the cluster plus 5 much closer galaxies. The indicies were calibrated against a nearby Landolt field - but this is my first time at calibrating this type of thing so I can't rule anything out.

Cheers

David

[higginsdj]

I have been informed elsewhere that it may have somethign to do with the location/mix of galaxy types! Elipticals are typically redder than Spirals so if the more distant galaxies are spirals then this is what my plot will show.

Cheers

David

[Hitchhiker]

Vespine makes a good point - red shift caused by the expansion of the Universe will shift the whole spectrum towards the red. Your B-V, R-I indices will reflect the intrinsic parameters of the galaxies, as the blue part of the spectrum will be shifted by the same amount as the red.

I think the only way to properly measure the red shift is to take a spectrum and determine the position of the lines for various elements against a 'reference' spectrum.

Thanks for posting this, I think the work you are doing is really interesting. Any chance you might post some of the data ?(here or on your website)

[rogerg]

Something I haven't understood about redshift is this:

You see the emittion lines from the distant object that have been shifted. But how do you know they have been shifted? Sure, we know that certain atoms emit certain lines in the spectrum (forgive me if my simplistic terminology is all wrong!) but how do we know the shifted emittion we are looking at is actually shifted and not different atoms emitting different bands in the spectrum?

err... if that makes sense.

Roger.

[pluck]

It's a pattern of lines which is shifted (not just one line). Each element has its own unique pattern (wavelengths) of emission (or absortption). It's easy, for example, to recognise the 'pattern' that Hydrogen emits, regardless of where it is on the spectrum.

[higginsdj]

Turns out it was redshift (higher values of blue in a B-V means redder - doh!)

For large scales, the redshift is very obvious and is referred to as photometric redshift in professional circles. It affects both the emission lines and the spectrum itself (remember redshift means the light waves are stretched - all the lightwaves) It's a few orders of magnitude less precise than measurement via spectral line shifts but doen properly it is accurate to 0.1 z (where z = redshift value).

A plot of the galaxies in B-V and R-I show a strong and steady shift from 0.01 z to -.15 z and I was event able to identify a second cluster of galaxies in the field. Abell 1631 at 0.05 z and a closer one at 0.015 z.

Cheers

David