|
Description
In the recent IBVS 5633, Roslund et al. give a list of new red variable stars. The work evidently was done in the early 1970s, but no effort was made to match the star-list with recent data available from modern on-line resources in this late publication. Johnson R magnitudes at maximum are shown for all stars and spectral classifications for about half. The authors warn that the magnitudes could have a systematic error, such that they are perhaps a magnitude too bright overall. Similarly, the coordinates, given to arcsecond precision, are noted as being poorly determined with "a simple, unpretentious plate-measuring machine of unproven accuracy." I confirm here both of these problems. I have gone through the list to recover the stars in modern catalogues. I matched the source-list in the Strasbourg VizieR utility against UCAC2 and 2MASS, searching a 20" radius and _assuming_ the brightest red star in the region is the variable (typical J-K>>1.0). There are no identification charts published, so misidentifications are possible. However, the candidates are all very similar in near-IR brightness (commonly 7<J<10), so it seemed the choices were limited. The original positions are indeed semi-accurate, being typically 5"-15" in error. I show improved coordinates in the table below. Nearly all thepositions come from 2MASS, but a few appear in UCAC2. That so few stars are present in UCAC2 is a first suggestion that the Roslund et al red magnitudes are much too bright compared to the standard system (either Johnson or Cousins R). I matched the corrected positions again in VizieR with GSC-2.2, USNO-B1.0, and the DENIS catalogue to obtain at least a "representative" V (or 'mv') magnitude, sometimes inferred from adjacent colors. Most of these come from the short-V plate data provided in GSC-2.2, where my experience with the same sorts of stars suggests the zero-point and scale are under good control. It was also easy to compare at the same time the Roslund R(max) magnitudes with the photo-red and I-band data listed in the three surveys just mentioned. This indicates the Roslund 'R' is at least one and more likely _two_ full magnitudes too bright even accounting for the large amplitudes of the variables---the zero-point is nearly that of the USNO-B1.0 or DENIS I-band magnitudes, rather than R. Finally, I matched the stars with the MSX6C and IRAS catalogues and also GCVS v4.2. Most stars have an MSX link, but probably the bright IR background close to the galactic center precluded many IRAS detections. I also searched each position in SIMBAD (3' search radius): the stars with Terzan links are shown, but the rest of the stars are not present in the database. Please note the following: where no reasonably bright red star appeared near the Roslund position on 2MASS J-band images, I used the identifications with the Terzan survey of the same region to obtain improved positions. This allowed all but one star to be recovered unambiguously.
|
