[vsnet-newvar 1503] Re: CGCS 2792

[Taichi Kato <tkato@kusastro.kyoto-u.ac.jp>]

Date: Sat, 20 Jul 2002 11:58:12 -0700 (MST)
From: Brian Skiff <Brian.Skiff@lowell.edu>
>>  Please explain your comments [about color terms for extremely red stars]
>> I know the original
>>  Johnson paper doesn't cover B-V >2, but Landolt has a few stars with B-V >2.

     Right.  But what happens at B-V = 4 ?  I can tell you:  the extrapolated
V magnitude will have veered away from "true" V (whatever that may be).
And this deviation from the standard system will be different for each
instrument.  In the case of the carbon stars this is because the passbands
of the system---be it an eye, a photographic detector, single-channel
photoelectric, or CCD--will have very small differences in shape that
will cause each to result in somewhat different values for the magnitudes
and colors.  In addition, the spectra of the carbon stars differs
significantly from that of the usual red standards (K or M giants), resulting
in what Sterken calls a "conformity error".  Yes, there are a couple of
warm carbon stars among the Landolt standards, but neither of them is
really stable, and their B-V colors are still only ~2.5, and in the (northern)
winter sky, so you can't use them now.
     So you end up having to do what amounts to differential photometry,
since there's no way to have everybody be on the same system for such stars.
If several observers got lots of coverage with significant overlap, then
afterwards you make an adjustment to each them to place the data on a
common system.  Or a single instrument could be used without adjustment,
but with the proviso that the data were more-or-less only on the 
instrumental system, or transformed closer(er) to the standard system by
the use of certain comparison stars.
     About the only way around the problem for carbon stars is to use
intermediate- or narrow-band filters that are centered on specific 
parts of the spectrum (say, a 'continuum' filter and a filter centered
on one of the fat C2 bands).  Calibratable, but then no standard V 
magnitudes would result.

\Brian

VSNET Home Page

Return to Daisaku Nogami

---

vsnet-adm@kusastro.kyoto-u.ac.jp