[vsnet-chat 3445] Re: Astrometry of Novae

[Brian Skiff <bas@lowell.edu>]

     Since no one's bit on this one, I'll offer my sole experience in the
matter.  If you are asking about position shifts in plates taken with the
same instrument, then such large offsets are possible, but would result from
either inappropriate observations or inadequate reductions.  Ordinarily one
would attempt (at least) for a bright object to have a fully-exposed but not
saturated image of the target, and hope that therre are enough fainter
reference stars to do the astrometry.  If you suspect there might be a
problem, then it is quite reasonable to include a magnitude-dependent term
in the position reductions.  Such an effect is clearly present in the GSC,
for instance, and the re-reduction for GSC v1.2 (the Space Telescope version,
not Bill Gray's GSC-ACT) includes this effect.  It shows up only on bright 
stars (roughly V < 8), so is not a problem for the fainter stars one usually
seeks in the GSC.
     More often, though, you have data (plates) taken with different
instruments, in which case things could be better or worse, given the variety
of errors that can crop up.  If the image scales of the two instruments are
different, or the reference nets differ, one can run into all sorts of
"funnies".  Or it can work right:  for one of the novae in Cassiopeia in
the 90s, I used a good early position to find a candidate progenitor on the
POSS-I prints, which I measured on our PDS machine (scanning microdensitometer)
used in semi-manual mode.  Later results from the Carlsberg meridian circle
were published, and their position during the eruption was well within our
mutual errors (mostly mine, working from paper prints).
     So I guess the answer to your question is "maybe", but one would want
to look into the details involved to decide case-by-case.

\Brian

VSNET Home Page

Return to Daisaku Nogami

---

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