measuring really long distances

February 17, 2006 11:10 PM

Do we have any measurements that confirm redshift for distant objects?

Reading this thread tickled a question that's been bothering me since high school. We have methods of measuring distances to nearby objects such as parallax measurements and measuring the brightness of cepheid variable stars. From that we get the redshift distance correlation. The part that has always bothered me is that as far as I know we extrapolate that correlation and use it as the only method of measuring the distance to remote objects. Do we have other methods for measuring distance to far away stuff? Is there enough evidence for the big bang that if we drop the observations based on redshift distance measurement that we can say that we live in an expanding universe and therefore the redshift distance correlation must be true?

posted by rdr to science & nature (10 comments total)

Yep! In fact, the name for the unit comes partially from parallax...Parsec. Cool stuff.
posted by SlyBevel at 11:40 PM on February 17, 2006

Hrm...got excited and didn't read the whole question.

Wikipedia offers a few explanations for red shift other than universal expansion here.
posted by SlyBevel at 11:45 PM on February 17, 2006

The Cosmic Microwave Background is considered to be pretty good evidence of the Big Bang. The uniformity of it seems to indicate that, at some point, everything was close together. The anisotropies in the background, as small as they are, are still very important (see the WMAP experiment).
posted by dsword at 11:48 PM on February 17, 2006

For the first part we use supernovae (type 1a specifically) to get us further than cepheids can, but using the same technique - using them as standardisable candles. There is recent work on using gamma ray bursts to do the same thing, but the data is a lot worse and in general astrophysicists are treating this GRB work more skeptically I think.

The distance to the CMB is determined by using our knowledge of how the galaxies are expanding with distance, which gives you a redshift-distance relation. As dsword said, the CMB is excellent evidence for the Big Bang, and there's other evidence too. If you change the redshift-distance relation then you can still use the anisotropies to examine things, but because they're now the wrong 'size' you end up thinking you live in a non-flat universe, so that the angular size you see them at matches up to the distance you think the CMB is at and the actual length scale you believe these anisotropies to have. You wouldn't stop believing in the Big Bang though, I don't think.

We'd still have other lines of enquiry that tell us the Big Bang is a plausible model - the nucleosynthesis predictions for example would still match up, but we'd perhaps not consider it quite as conclusive as what we have from redshift measurements, and we'd certainly not be doing any 'precision cosmology'.

So - other methods for determining distances? Well, no, I don't think we do have other methods than the ones you've probably come across, at least not any sufficiently accurate. What we do have are other methods that are consistent with the distance scales we use (we might say that more distant galaxies look smaller for example), but these won't really pin things down to the necessary accuracy (we don't know the physical size of distant galaxies to really nail things down with).
posted by edd at 4:18 AM on February 18, 2006

[Extra linkage: an excellent Big Bang FAQ and a distance scale FAQ which may be handy for other readers of the question to get background. They're from TalkOrigins.org so sometimes spend time dealing with creationist complaints and alternative (ie crackpot) ideas, but they're amongst the best articles I've seen for explaining these things]
posted by edd at 4:25 AM on February 18, 2006

Consider also that if they look at a nearby galaxy (and there will be some that are oriented the right way), there's going to be a red shift on the side rotating away, and a blue shift on the side rotating towards. The size of a galaxy tells them about how fast it must be rotating, so that should let them verify that their redshift calculations are correct. (The galaxy would have to be close, or it would already be moving so fast that the relatively minor shift from orbital velocity would get lost n the noise.)

All of these measurements have to be self-supporting to hang together; if our figures for blue and redshift are wrong, then something would be inconsistent in the data we were collecting. There are strong arguments about the nature of gravity over very large distances (which are related to the arguments over dark matter and energy), but essentially none over shift measurements.

As far as I know, the measurements are _extremely_ consistent, which strongly suggests that we have it about right.
posted by Malor at 6:19 AM on February 18, 2006

For certain distances anstronomers can triangulate the distance to obects by using the Earth at two extremes of its orbit, or even two distant points on the Earth as two of the three points making up the triangle. Here's the How Stuff Works link.
posted by princelyfox at 10:48 AM on February 18, 2006

The parallax methods described by princelyfox won't get you out far enough to get to distances where there is an observable cosmological redshift. See the distance scales FAQ I posted above - you need to go beyond the nearest galaxies to start developing a redshift-distance relation - the reason being that bound systems such as the galaxy and the local group of galaxies do not expand as the universe as a whole expands.
posted by edd at 11:48 AM on February 18, 2006

Well, from what I understand, you can tell if something is red-shifted based on specific light frequencies you'd expect to see from particular elements, like hydrogen going through fusion or otherwise getting hot.

If the question is 'how do we know that a particular frequency shift really corresponds to distance' then I don't know, but in order for that to be wrong I think the speed of light would need to change over very long distances, which is possible, but probably not likely.
posted by delmoi at 3:02 PM on February 18, 2006

but in order for that to be wrong I think the speed of light would need to change over very long distances,

Not really. Something can be close and also moving away from you very fast.

Essentially edd is right. The longest standard candle we have is the Type 1a Supernovae which I believe give us a range out to billions of light years. They seem to all follow the same luminosity profile as seen by looking at them when they occur in nearer galaxies and comparing them to other known standard candles.

How do we know the redshift distance relationship still works for galaxies even farther away? We don't. But there are some strong reasons to believe its still true. One is, as edd pointed out, that smaller, dimmer galaxies all tend to have the highest redshift. Another is that we can do Galactic redshift surveys and they show that if we assume the redshift/distance relationship for distant galaxies we get a galactic distribution that, on the whole, looks the same as that for the galaxies within the supernova boundary. That is, it all seems to hold together.

I would be remiss however in not mentioning quasars. These are extremely high-redshift objects that are also really bright. If you stick to the redshift relationship then they are really far and the source of their immense power is a puzzle. However, there is a significant number of scientists, most prominently Hans Arp, who believe they are exceptions to the redshift relationship.

Just the fact that there is controversy here should tell you that in some sense the true distance of high redshift objects relies as much on induction as it does in deduction.
posted by vacapinta at 5:53 PM on February 18, 2006