How new stars data were extracted from the ascc catalog
How new stars data were extracted from the ascc catalog
I want first to thank Alexis Brandeker, an astronomer. Without his help in the development of the discrimination procedure for Tycho stars, his knowledge in astronomy, I could not success in this project...
The All-Sky Compiled Catalogue of 2501968 stars (ASCC-2.5) with the limiting magnitude V=12-14 is a result of a merging of star lists from present day large high-precision catalogues from space (Hipparcos-Tycho family catalogues: Hipparcos main catalogue including Multiple System Annex <I/239>, Tycho-1 <I/239>, Tycho-2 <I/259>, ACT-RC <I/246>, TRC <I/250>) and ground-based (PPM-N <I/146>, PPM-S <I/193>,PPM-add <I/208>, CMC11 <I/256>) observations and reduction to standard systems of corresponding stellar data. Main stellar data presented in the ASCC-2.5 are the equatorial coordinates (J2000, epoch 1991.25), proper motions in the Hipparcos system, B and/or V stellar magnitudes in the Johnson system. Additionally for some stars are given trigonometric parallaxes, spectral classes in the MK or HD system, multiplicity and variability flags, Hipparcos, Tycho-2, HD, DM designations.
Well, the problem was to extract some "good" stars out of the 2454555 Tycho stars of the ASCC catalog where parallax values are often unusable, either because they are negative, or lesser than their corresponding parallax errors values.
Thus, only 28059 (V 2.0 A) stars with "good" parallax values could be directly extracted from the catalog.
The other idea was to use the B-V color index of a star, to assume its luminosity class is either III (giants) or V (dwarf - a main sequence star like the sun), with the help of discriminating polynomial functions, and then to compute its absolute magnitude and its estimated distance.
Our procedure is still not perfect and make some assumptions on the luminosity class of a star, based on statistical results (see the following first HR diagram). Thus the distances that are estimated for these new stars, can have up to 50% relative errors.
A HR diagram generated from 26165 Hipparcos stars of all luminosity classes.
Class III : 27 %
Class V : 59 %

A HR diagram generated from 15364 Class V and 5708 Class III Hipparcos stars.
P3 and P5 are our polynomial fitting functions that are used in our discriminating procedure.

A Tycho star is then accepted by our procedure, if it has no Hip number (quite all Hipparcos stars are already in the Celestia stars database), if its apparent magnitude (appMag), parallax (Par), parallax error (ePar), B-V color index (BV), error on B-V color index (eBV), RA and declination fields are valid,and if one of the three conditions is satisfied :
1) | Par | <= 0.0 and | eBV | < 0.15 then we assume the star is of class V if | BV | < 0.7 (see the HR diagram above)
2) | Par | > 0.0 and | ePar | / | Par | < 0.35 (stars with "good" parallax value)
3) | Par | > 0.0 and | eBV | < 0.15 then we assume the star is of class V if | BV | < 0.7 (see the HR diagram above)
In cases (1) or (3) when | BV | >= 0.7, we compute then an estimated absolute magnitude and with P3 and P5 polynomial fitting functions, we deduce the luminosity class (III or V) of the star.
Of course, this procedure could be improved.
Thus 887643 from the 2454555 Tycho stars were accepted with these criterias. Here is the output of the extraction procedure for V 2.1 A :
Number of Tycho stars analysed : 2454555
Number of Tycho stars accepted : 887643
Number of Tycho stars rejected : 1566912
Number of Tycho stars with Hip numbers rejected : 124235
Number of Tycho stars rejected through BV out of range : 30674
Number of Tycho stars accepted through acceptable parallax value : 5604
Number of Tycho stars accepted through acceptable B-V error value : 882039
Number of Tycho stars accepted through acceptable B-V error value with initial neg. parallax : 494514
Number of Tycho stars accepted of Luminosity Class III : 185036
Number of Tycho stars accepted of Luminosity Class V : 702607
Number of Tycho stars with appMag < 6.0 : 6
Number of Tycho stars with 6.0 <= appMag < 7.0 : 98
Number of Tycho stars with 7.0 <= appMag < 8.0 : 4405
Number of Tycho stars with 8.0 <= appMag < 9.0 : 39644
Number of Tycho stars with 9.0 <= appMag < 10.0 : 188776
Number of Tycho stars with 10.0 <= appMag < 11.0 : 444736
Number of Tycho stars with 11.0 <= appMag < 12.0 : 209690
Number of Tycho stars with 12.0 <= appMag < 13.0 : 288
Number of Tycho stars with 13.0 <= appMag : 0
Number of Tycho stars with distance <= 25 pc : 0
Number of Tycho stars with 25 < distance < 50 pc : 44033
Number of Tycho stars with 50 <= distance < 100 pc : 43500
Number of Tycho stars with 100 <= distance < 500 pc : 520551
Number of Tycho stars with 500 <= distance < 1000 pc : 250659
Number of Tycho stars with 1000 <= distance < 2000 pc : 27177
Number of Tycho stars with 2000 pc <= distance : 1723
Number of Hipparcos stars dropped : 903
Number of Hipparcos stars with parallax set to limit : 53
Stars processed: 1017366 Number dropped: 981 number dubious: 1036 .
Total stars count in stars.dat : 1016385
DIFFERENCE BETWEEN V 2.1 A and V 2.1 C
I have made two versions of the V 2.1 new stars database.
In the version 2.1 A of 1016385 stars, the extraction parameters are set to :
parallax relative error < 35% and B-V error < 0.15.
In the version 2.1 C of 2072873 stars, the extraction parameters are set to :
parallax relative error < 45% and B-V error < 0.35.
Thus V 2.1 C contains more stars than V 2.1 A but with lower precision on distances.
How Tycho stars are labelled in Celestia
A certain amount of Tycho stars have a HD number and can be searched in Celestia with this number (ex : HD 169236)
This HD number comes either from the original HD catalog (range 1-225300) or from its extensions HDE (range 225301-272150) and HDEC (range 272151-359083).
A Tycho star number has the form : TYC Tyc1-Tyc2-Tyc3. Ex : TYC 8745-250-1
To avoid conflicts with real and existing Hipparcos numbers in Celestia, and to allow searching for Tyc stars in Celestia, I have used the following formula :
Fictious HIP number = Tyc1 + 10^4 * Tyc2 + 10^9 * Tyc 3 (or abbbbbcccc where a = Tyc3; bbbbb = Tyc2 and cccc = Tyc1)
Thus our example TYC 8745-250-1 will have the following fictious HIP identificator in Celestia : HIP 1002508745.
All fictious HIP numbers for new Tycho stars are therefore coded on 10 digits and greater than 1 000 000 000.
may 2003 Pascal Hartmann