Accuracy of the relative positions of the Virgo galaxies.

[dave1235]

I set up my demo.cel script to fly through the heart of the Virgo Cluster on its way to planet Earth. I was wondering how much stock I should be putting into what I find when it comes to the positions of the different galaxies in the cluster relative to each other. Most websites just list the distances of the central Virgo Cluster Galaxies as being the same large round number. There is however a difference between how far the Cluster center is away from us and what the galaxies positions are relative to each other. Again could you give me a sense of the accuracy of what I?€™m looking at. I?€™m thinking mainly of the nearby Messier Objects.

Thanks
David

( Fridger is expecting this post) If he doesn't have time to answer could someone give me a ruff idea. Its something I'm sure a lot of people would like to know.

[Cham]

Fridger is fine tuning his galactic catalog. Currently, there are some distance errors and inaccuracies related to uncertainties. The next revision, to be published soon (?), should be much better.

[t00fri]

I set up my demo.cel script to fly through the heart of the Virgo Cluster on its way to planet Earth. I was wondering how much stock I should be putting into what I find when it comes to the positions of the different galaxies in the cluster relative to each other. Most websites just list the distances of the central Virgo Cluster Galaxies as being the same large round number. There is however a difference between how far the Cluster center is away from us and what the galaxies positions are relative to each other. Again could you give me a sense of the accuracy of what I?€™m looking at. I?€™m thinking mainly of the nearby Messier Objects.

Thanks
David

( Fridger is expecting this post) If he doesn't have time to answer could someone give me a ruff idea. Its something I'm sure a lot of people would like to know.

David,

being a theoretical physicist (with vast research expertise in particle and astro-particle physics and cosmology) , the concept of measurement errors is of intrinsic importance to me. Along with this, goes the requirement that all data used by Celestia MUST be transparently documented. As concerns my strategy of systematically filling the Celestia Universe with MANY deep-space objects, this has been largely achieved, however. The transparency of documented modifications to the Hipparcos star database will also be improved in the near future (see my thread on stars.txt in
http://www.celestiaproject.net/forum/viewtopic.php?t=10829)

++++++++++++++++++++
However many Windows users may not be aware of the complete existing documentation for galaxies and binary orbits, since the binary version distributed by Chris does not include it. I suppose we shall change this in future versions!
++++++++++++++++++++

Here comes a summary of the status of documentation for my galaxy database deepsky.dsc (and analogously my databases for visual and spectroscopic binary orbits, visualbins.stc and spectbins.stc)

1) The database is generated for Celestia without any human interference directly from respected published scientific catalogs by means of a corresponding PERL script (of meanwhile 950 lines). These catalogs are always to be found on one of the professional catalog servers of the global astronomy database, e.g. in Europe

http://cdsweb.u-strasbg.fr/cats/Cats.htx

2) The actual catalogs used along with the citation of the respective scientific journal and/or the WEB site are always cited at the top of my database files. Have a look!

Code: Select all

# Revised NGC and IC Catalog, Wolfgang Steinicke, April 5, 2005
# http://www.ngcic.org/steinicke/default.htm
#
# Augmented by
# Revised 3rd Reference Catalog of Bright Galaxies (RC3,VII/155)
# http://cdsweb.u-strasbg.fr/viz-bin/Cat?VII/155
#
# Augmented by
# Mark III Catalog of Galaxy Peculiar Velocities (Willick+ 1997,VII/198)
# http://cdsweb.u-strasbg.fr/viz-bin/Cat?VII/198#sRM2.18
#
# Augmented by distances from
# The SBF Survey of Galaxy Distances. IV.
# SBF Magnitudes, Colors, and Distances,
# J.L. Tonry et al., Astrophys J 546, 681 (2001)
#
# Augmented by distances from
# Compilation of "200 Brightest Galaxies"
# http://www.anzwers.org/free/universe/galax200.html
#
# Using today's Hubble constant = 72 [km/sec/Mpc]  (WMAP 2005)
#
# Abreviations for various distance methods used:
# S= SBF (Surface Brightness Fluctuations), T-F= Tully-Fischer
# V = rad. velocity in CMB frame & Hubble law
# C=Cepheids, P=photometric, N(G)=planetary nebula (globular cluster) luminosity function
#
# Adapted for Celestia with Perl script: deepsky.pl Revision: 1.20
# Processed 2006-1-3 2 2 0 21:19:26 UTC
#
# by Dr. Fridger Schrempp, fridger.schrempp@desy.de
# ------------------------------------------------------


My PERL script merges the information from the various individual catalogs in a reproducable manner!

For the forthcoming vastly improved deepsky.dsc the citations in addition include:

Code: Select all

# Augmented by averages of distance entries from
# NED-1D: NASA/IPAC Extragalactic Database of Distances
# http://nedwww.ipac.caltech.edu/level5/NED1D/
#

# Augmented by redshifts-> distances from
# VII/193 The CfA Redshift Catalogue, Version June 1995  (Huchra+ 1995)
# http://cdsweb.u-strasbg.fr/viz-bin/Cat?VII/193
#

along with the updated NGC/IC database

# Revised NGC and IC Catalog, Wolfgang Steinicke, [b]January 6, 2006[/b]
# http://www.ngcic.org/steinicke/default.htm
#


+++++++++++++++++++++++++
All the inherent uncertainties of my database can therefore be found in the cited papers!
+++++++++++++++++++++++++


3) My PERL script generates ALL the input required for the galaxy visualization in Celestia, exclusively from the catalog data, usually involving significant intermediary calculations, however. One typical example is the calculation of the sizes and spacial orientations of the galaxies in the Celestia frame of reference, using transformations encoded by means of quaternions in my script.

4) In a fair number of cases, the existing catalog information is not sufficient and additional astro-physical model assumptions are required for achieving the display. Sometimes also the catalogs contain obvious errors requiring modification, of course

+++++++++++++++++
All corresponding action is documented in form of my well-commented PERL scripts that are ALWAYS part of the Celestia source distribution, available from the official Sourceforge server:
+++++++++++++++++

http://sourceforge.net/projects/celestia

or for the development versions in the respective CVS archive.

My script deepsky.pl generating deepsky.dsc is located in the source directory

celestia/src/tools/galaxies

while the PERL scripts responsible for the binary orbits are found in

celestia/src/tools/binaries

As a crucial feature, PERL scripts are human readable also for people unaware of the PERL scripting language!
The inputs to all scripts are ALWAYS the UNMODIFIED, published catalog data!

+++++++++++++++
In summary: every single deep-space object along with its inherent uncertainties is completely documented
+++++++++++++++

I hope this has cleared up some important issues about the documentation of deep-space data and their errors in Celestia!

Bye Fridger

[t00fri]

Fridger is fine tuning his galactic catalog. Currently, there are some distance errors and inaccuracies related to uncertainties. The next revision, to be published soon (?), should be much better.

Unfortunately, this is a comparatively slow process, however, since many of the required distances for the NGC/IC galaxies are still virtually unknown.

Here is the latest accounting for my complete "revised NGC/IC" galaxy catalog. The counts are always output by my PERL script after generating deepsky.dsc:

Code: Select all

 1074 Averaged distances from NED-1D Distance Data Base
  289 SBF galaxy distance data
  544 TF elliptical galaxy distance data
 2944 TF spiral galaxy distance data
  200 B200 spiral galaxy distance data
 7302 zcat redshift data
16257 RC3 galaxy velocity data (CMB system)

Total:                  10611
with good distance data: 7387

NED-1D distances:         720
SBF distances:              2
T-F distances:            977
B200 distances:            34
Hubble Law (v_CMB, RC3): 4363
Hubble Law (ZCAT):       1291


Since the new NED-1D data base tends to contain several independent distance measurements per galaxy (with the used methods indicated) , I have chosen to average all distance values in NED-1D for each galaxy.

For the fairly large remaining number of galaxies where distance (or at least velocity or redshift) measurements are not explicitly available I have implemented the following rough estimate:

I assumed the absolute galaxy magnitudes to be a constant equal to the /peak/ value of the known distribution from the accurate SBF-distance measurements (i.e. m - M for 289 SBF-galaxies along with separately measured m's). The peaks are quite SHARP and differ somewhat for spirals and ellipticals +irr's. This I took into account.

The actual distributions of the absolute magnitude (M_abs) of the SBF- galaxy data look like this,

blue=spiral data, red = ellipticals & irr's and area normalized to ONE,

whence I extracted the following peak values by computer

Code: Select all

<M_abs_ell> = -20.133;
<M_abs_spir> = -19.07; 


These were used in all previous versions of deepsky.dsc as clearly documented in the corresponding deepsky.pl PERL script.

In parallel, it was/is always indicated in deepsky.dsc when this uncertain distance estimate was used, like so:

Code: Select all

Distance   5.339e+08  # distance uncertain!


instead of e.g.

Code: Select all

Distance   4.501e+07  # method: NED-1D average


Since the new NED-1D project will be continuously updated and notably extended, there is good hope that the situation will considerably improve quite soon.

Let me quote from the NED-1D Introduction
http://nedwww.ipac.caltech.edu/level5/NED1D/intro.html

We anticipate a primary need for NED-1D to provide a new, minimum baseline for the number of accurate distances known, and to be ready just in time to begin assimilating the many thousands and soon tens of thousands of accurate distances that will inevitably be published after launch of NASA's next-generation, James Webb Space Telescope mission in 2013 (see, Gardner et al. 2006). In addition to maintaining up-to-date information, via future updates as customer specified, we anticipate user needs for additional versions of NED-1D, as part of an overall NED-D system. A complete system will provide additional distance information not currently included, and incorporate other distances, not now provided. Areas of future interest have so far been categorized into the following, potential versions.

Bye Fridger

[dave1235]

Thanks, I read it all twice but?€¦.
While all that information is important and should help put Celestia on a firm scientific basis in the minds of astronomers

You say

All the inherent uncertainties of my database can therefore be found in the cited papers!

and I respect that, but I'm not a hardcore astronomer.

My experience with star distances is this... I spend wads of time memorizing the distances of several stars in a table so that I can recall them later while out star gazing only to find that the distances cited were more like guesses.
I understand the difficulty estimating stellar distances. The thing is it would be very useful for me and I'm sure for many others who want to know how "real" what they?€™re looking at is to have some sort of error indicator that they can turn on.

I know you guys have many more ideas you can implement then time to do it, and I guess it all comes down to personal priority. And I would also guess that handling and displaying another type of data would be a lot of work. But I?€™m really just looking for a general idea.

And again I?€™m more interested in their relative positions to one another than their distances from our galaxy.

But then again I appreciate how that might not be possible.

Thanks for your time.

David

[t00fri]

David,

it's really not very wise to try and memorize those values. That's what we have computers for!

I have even been joking elsewhere in the forum about incorporating a graphical means for displaying uncertainties. But that's not easy! It's particularly hard in case of galaxy distances because the uncertainties are mainly what physicists call systematic errors in contrast e.g. to statistical ones.

Systematic errors are typically attached to the method of measuring data and these type of errors unfortunately do not improve by measuring MORE such objects. We use about ~10 different methods of determining galaxy distances, each with its own systematic error range. It is even hardly possible to compare the uncertainties of these different methods.
Averaging of systematic errors is another problem area...

Since I know more than what is healthy about treatments of errors in physics, so far I prefer to skip error displays altogether. This still seems to be the more transparent compromise. Remember: Bad error estimates are worse than NO errors!

And since you say that you are not a professional astronomer, that should not hurt you too much. If you are getting over-curious => study the publications.

Bye Fridger

[dave1235]

All points well taken.

Might I add that given that you...

use about ~10 different methods of determining galaxy distances

There is the quick and dirty option of displaying the distance for each star or galaxy once for each time it?€™s listed in a different catalog, the result would be a kind of probability group in space. I don?€™t know whether that would be misleading or not but it would get the message across to the professional astronomers that time and effort is being taken to collect and display the positions of the stars and galaxies as accurately as anyone else is going to, given the current data.

Anyway thanks for your time.

David

(By the way Fridger you might choose to copy and paste a modified version of your above posts in the FAQ section under something like?€¦?€