[vsnet-newvar 1715] re re MisV1147 Spectral Type

[jg@jgws.freeisp.co.uk]

Doug West noted :-

> For observation number 1
> Q = (U-B)obs -  0.72*(B-V)obs = 0.087 - 0.72*(0.814) = -0.5
> Which equates to spectral type B4.5
> 
> For observation number 2
> Q = (U-B)obs -  0.72*(B-V)obs = 0.044 - 0.72*(0.933) = -0.63
> Which equates to spectral type B2.5

Or B5 ish ;^)

Observation 1 at V = 13.4 and Observation 2 at V = 14.9 are more or less
max and min for this object.

Such a deep minimum for two stars in a binary of such equal type is
unlikely.  Similarly the lightcurve would be expected to be quasi-EW not
quasi-EA if there were two B type stars orbiting each other.  A B2.5
star, ostensibly brighter than a B4.5 one, causing an eclipse! ;^)  This
argues against binarity then, at least in the context of photometrically
derived spectral class.

Anyway, as noted in the past by AH, the thing gets redder except for in
U-B during "light weakenings".  Now, there is no qualitative difference
in B-V, V-Rc, Rc-Ic, B-Ic, but there is between U-B and any of these. 
This is why you see diagnostic plots of groups of stars showing U-B
versus B-V, but none similar for anything like B-V versus V-Ic etc. In
fact, this goes all the way down to K.  The next passband colours you'll
see plotted against each other in Johnson-Cousin's systems as being
qualitatively diagnostic are J-K versus K-L.  I'm pretty sure the U-B
difference is because it is based around the Balmer Discontinuity, not
based on a continuum with a Wien's Law relationship like B-V, but have
no idea what is happening at the K / L "boundary". So the other colour
indices derivable here will add nothing.


I feel a forward scattering of U-B wavelengths (similar to what happens
in reflection nebulae) in tandem with normal reddening (de-blueing) in
B-Ic passbands may say something about the nature of the circumstellar
matter for this object (particle size, gas/dust ratio, etc), but I
cannot readily find _quantitative_ details on this topic.


It is certain that any optical polarimetric measure of this object would
be highly diagnostic.  If it's a YSO Ae/Be star there'll be
polarisation.  It's a bit faint for amateurs to manage this, even if
they had such a filter to hand.  Polarisation during eclipse events
would be especially interesting with respect to the identity of the
eclipsor.


Cheers

John Greaves

VSNET Home Page

Return to Daisaku Nogami

---

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