Galaxy Distances - Catalog of 1791 distances (Tully et al, 2008)

[madbadgalaxyman (Robert)]

A recent master catalog of 1791 galaxy distances is that of R.B. Tully et al., 2008, ApJ, Vol. 676, p.184

The catalog itself is not fully included in this paper, and it is referred to in this paper as "Catalog of Galaxy Distances"

However, when actually accessing the catalog data at the CDS website, the catalog has this awkward name:
"Peculiar Motion Away from The Local Void (Tully+ 2008)"

(which is odd, as this title states one possible purpose of using this catalog, rather than what the catalog actually contains!!)

The galaxy distances in this catalog are composed of the following:
(1) 1030 individual galaxy distances derived by means of the Tully-Fisher relation, which are accurate at about the 20 percent level; this accuracy is good enough for finding the distance of a Galaxy Group or Galaxy Cluster (if you average the distance estimates for all the cluster members, you end up with a reasonably reliable distance for a cluster of galaxies)
(2) The majority of the other galaxy distances in this catalog were determined using the more reliable techniques of : Cepheid, SBF and TRGB methods. These distances are nominally accurate to +/- 10 percent.

You will note that the distances of galaxies that are given in this catalog are given as distance moduli, rather than as actual physical distances in Megaparsecs(millions of parsecs). However, if need be, I can give you a simple formula (using college/high-school algebra) that quickly converts the distance modulus of a galaxy into its actual distance. I can also help with "de-jargonizing" some of the arcane symbols used in this catalog!!
(but don't ask me how to convert the heliocentric recession velocity of a galaxy into the "Local Sheet of galaxies" reference frame!)

How to access the catalog of 1791 galaxy distances:

(1) Access the website of the CDS data archive:
http://cdsarc.u-strasbg.fr
(2) Click on "explore and look for catalogues"
(3) Write "Tully" in the search box, and press ENTER
(4) A list of various galaxy catalogs and data tables comes up, all of them with R.B.Tully as one of the authors.
(5) Find, in this list of galaxy catalogs, the one with this title: "Peculiar Motion Away from The Local Void(Tully+ 2008)"
(6) Click on "Vizier" at the right of this catalog name, to begin the process of displaying the catalog data.
(7) On the next webpage which comes up, click on "J/ApJ/676/184/table1" in order to bring up the page on which you will select which columns of data to display.
[ You can also select the number of rows of data you wish to display;there are over 1000 rows in this catalog of galaxy distances. ]
(8) Then, a page comes up in which you use check boxes to decide which of the dozens of columns in this catalog you actually want to display. At an absolute minimum, I suggest displaying the: record number, galaxy coordinates, galaxy ID, galaxy recession velocity, and the galaxy distance measurements (distance modulus)
(9) Click on "Submit" to bring up a viewable and printable version of this catalog of distances of various galaxies.

Enjoy!!

Galaxy Distance Measurement techniques (fortunately) use simple algebra of the sort that many of us learnt at school or college. A very thorough and relatively easy review paper is:
http://ned.ipac.caltech.edu/level5/J..._abstract.html

[Weltevreden SA (Dana)]

As an addendum to Robert's excellent explanation, the story of how Tully & Fisher came up with it is pretty fascinating in its own right.

I hope this isn''t off-topic, but one of the key terms mentioned early in the Tully-Fisher article is the 'virial theorem'. Understanding what that means is important because it so often comes up and in so many contexts—galaxies, globulars, cluster formation, the ISM, you name it. I've yet to read an explanation of the word 'virial' and the related terms that go with it—'theorem', 'radius', 'mass', and so on—written by someone whose mother tongue isn't partial differential equations.

Any chance you'd like to take a crack at it, Robert? You have a great ability to turn the complex into the lucid.

[madbadgalaxyman (Robert)]

Originally Posted by Weltevreden SA

(....)the story of how Tully & Fisher came up with it is pretty fascinating in its own right (....)
(.......) one of the key terms mentioned early in the Tully-Fisher article is the 'virial theorem'. Understanding what that means is important because it so often comes up and in so many contexts

I have considered trying to explain the virial theorem for the enlightenment of our IIS members, as a few IIS people have said that dark matter is only an uncertain assumption which was put in so as to be an "unknown type of matter" which can make the equations turn out right. This is absolutely an incorrect view.
In fact, it has been plain since the work of Fritz Zwicky in the 1930s that the virial theorem shows that galaxy clusters would fly apart if there were no gravitating dark matter to hold them together. This conclusion comes from well-established data and simple existing physics; though obviously one can speculate as to some other reason for the effect! As a leading dark matter theorist told me "Firstly, you have to go where the existing data and the existing physics leads, so the most parsimonious interpretation is that there are large quantities of additional gravitating matter that do not emit much radiation of any sort."

I might try to explain the virial theorem, but don’t hold your breath waiting for me to do it……as I just downloaded all the relevant papers on the Tully-Fisher relation, and I am very busy trying to absorb them!

The Tully-Fisher relation has long been considered as a promising method for an accurate determination of galaxy distances based on only one simple observable; the maximum rotational velocity of a disk galaxy, as derived from the width of the 21 cm line coming from the HI within a disk galaxy.
The intrinsic linewidth (caused by random velocities) of the HI line (which comes from cold neutral atomic hydrogen) is quite narrow within the disk components of non-dwarf spiral galaxies, so you just have to measure the rotational width of this spectral line, that originates in a quick-rotating disk structure, in order to derive a rotational velocity for a spiral galaxy or a galaxy of type S0 .

However, it has turned out to be difficult to make the necessary correction of the observed line-of-sight velocity near the edge of a Spiral Galaxy to the actual rotation velocity of a galaxy, as there are several possible errors in deriving the inclination (that is, the orientation from the observer’s point of view) of a rotating disk.(after all, disk galaxies can’t be measured with a ruler!!)
For instance real spiral galaxies are often not exactly circular and can have significant intrinsic ellipticity (the actual shape of the galaxy in real 3-D space is somewhat oval). To make things even more complicated, some specific spiral galaxies do not even have a single orientation, as they are known to be composed of several fairly-distinct disks or annuli, at some angle to one another.
(A better solution for figuring out the precise orientation of a disk galaxy is to measure the line-of-sight velocity at every point over the surface of the image, which is now possible with some types of spectrographs. Such a velocity field can also show if there is a bend or 'warp' in the disk of a spiral galaxy , at some radius.)

A further problem with the Tully-Fisher method is that the observed magnitude of a galaxy (how much light actually reaches the observer) depends in part on the amount of ‘internal’ extinction of the galaxy’s light from the screen of interstellar dust that is within it, and there are various conflicting mathematical prescriptions for calculating internal extinction.

Sadly, despite over 20 years of work, the scatter in the Tully-Fisher relation remains large; about 0.4 magnitudes in luminosity, which is equivalent to a 20 percent error in distance. And this already large error does not include an additional baseline 'calibration' error in the entire Cosmological Distance Ladder...... which has its origin in the uncertainty about the precise distance of the LMC.

Still, one must remember that the Cepheid Variable method of distance estimation, which had its debut with Henrietta Leavitt in 1912, took most of the rest of the 20th century to become anything more than a rough and ready estimator of galaxy distances.

These calibrations of the T-F relation are still commonly used in galaxy distance estimation:
(1) Some of R.Brent Tully's more recent calibrations of the Tully-Fisher relation can be found in:
2000, ApJ, Vol.533, 744
2008, ApJ, Vol.676, 184
(2) Tully's opinion on the still controversial issue of extinction within spiral galaxies:
1998, AJ, Vol.115, 2264
(3) A calibration of the T-F relation based on Cepheid distances: 2000, ApJ, Vol.529, 688

Added, in an edit:
The T-F method, rather disappointingly, does not really yield a distance estimate for a galaxy that is more accurate than the distance that can be estimated from the recession velocity of that galaxy. However, observed recession velocities of galaxies can be strongly affected by the individual motions of galaxies (typically up to 300 km/s, for those galaxies outside of galaxy clusters, within a redshift of 6000 km/s).
Therefore, when the "learned astronomer" is trying to figure out how far away a specific galaxy is, it is most useful to have both a T-F distance and also a distance which is derived from the recession velocity of the galaxy. (I am still working on the distance and luminosity of SN 2013aa !!)

[CarlJoseph (Carl)]

Hi,

This is a very timely discussion for me. I'm currently writing up my project in the Tully-Fisher relation for my Swinburne degree (SAO). Am using data from HIPASS which was a survey of Southern Hemisphere galaxies done with the Parkes radio telescope.

If people are interested I would happily post more about this topic once I've finished my project. I can't post my project paper itself due to Swinburne's policy on aiding plagiarism, but will send it to people who might be interested.

Suffice to say, distances are a precarious thing. Look up the "distance ladder" and you will soon realise that it might be safer to just not climb so high!

Cheers,
Af.

[madbadgalaxyman (Robert)]

Originally Posted by Afro Boy

Hi,

This is a very timely discussion for me. I'm currently writing up my project in the Tully-Fisher relation for my Swinburne degree (SAO). Am using data from HIPASS which was a survey of Southern Hemisphere galaxies done with the Parkes radio telescope.

If

Cheers,
Af.

All in all, TF has been a disappointment, with no progress in distance accuracy over the last 20 years. It is no more accurate than deriving a galaxy distance from a recession velocity, and sometimes less accurate.
TF estimates still have value as part of a mean distance, whether a mean of several different distance estimates for the same galaxy, or by getting TF distances for multiple galaxies within the same cluster and then averaging these distance estimates. TF is mainly used, these days, for deriving the peculiar (non-cosmological) velocities of clusters of galaxies.
The best methods of distance estimation (TRGB ; Cepheids ; SBF) have an error of about 10 percent, plus an additional 5-10 percent calibration error due to uncertainty about the distance of the LMC

Still, we are in a better position than the situation during the "Hubble wars" between Alan Sandage and Gerard de Vaucouleurs, when there was a factor of two uncertainty in galaxy distances!

There could still be serious errors in the distances and luminosities of our 'baseline' calibrators within our own galaxy, due to the very limited current range of the parallax method of distance estimation.

[CarlJoseph (Carl)]

You're right, it has been somewhat disappointing, although it's also been found incredibly difficult to calibrate. For example, different morphologies lead to different relationships. There is also different scatter present in different colour bands so your choice of band will also impact your result.

It appears to function better at working out distances to clusters as you mentioned. There is also a recent paper by Source et al. (2013, ApJ, volume 94) which calibrates the TFR for the mid-infrared band. This band appears to have much less scatter and according to their research has provided slightly better results. There might still be hope yet!

Originally Posted by madbadgalaxyman

There could still be serious errors in the distances and luminosities of our 'baseline' calibrators within our own galaxy, due to the very limited current range of the parallax method of distance estimation.

This is very true. The Gaia mission should help improve that (http://www.esa.int/Our_Activities/Sp.../Gaia_overview).

With the Hubble constant, there is also quite a bit of variation. I've seen recent studies showing results from 76 down to 67 km/s/Mpc (Planck 2013 results. I. Overview of products and scientific results, A&A, 2013).

All very interesting.

Cheers,
Af.

[Weltevreden SA (Dana)]

Af, could you elaborate on your comment of the 26th, 'There is also different scatter present in different colour bands'? Does 'different' mean 'proportional across the arcsec region of a specific observation run,' which implies object- or field-related results? Or rather 'different equipment produces slightly different results' which is setup and data-analysis related? Differential band scatter raises issues about the accuracy of studies where comparative metallicity is crucial, e.g., Norris, Yong et.al 2012? I would be interested in any links quantifying the phenomenon that you could provide. =Dana

[CarlJoseph (Carl)]

Hi Dana,

Thanks for the question. I think I may have confused things a little ... By different colour-bands, I mean observing at different wavelengths using filters.

You can view a galaxy (or a star for that matter) through different filters, e.g. blue, red, infrared, etc. Different wavelengths of light are impacted by the interstellar medium and galactic dust in different ways. Blue light for example is dimmed more than red, and infrared light. We can correct for it pretty well but it's not perfect.

Because of that, magnitudes of galaxies on the Tully-Fisher line show different levels of scatter (distances from the line) for different wavelengths. Blue and Red show more scatter than near-infrared measurements. So longer wavelengths in the infrared range seem to show a stronger/closer relation to the Tully-Fisher line.

Hope this helps.

Cheers,
Af.

[Weltevreden SA (Dana)]

Thanks for your clarification, Af. It stands to reason that IR scatter would be <R <V <B <U, etc. I wanted to clarify how you were using the term. I'd be interested in reading your Swinburne HIPASS paper when done. In summary form if need be because of the SAO plagiarism concern. =Thanks, Dana