Re: comp stars, software, and such

From: Joe Patterson <*jop_at_astro.columbia.edu*> Date: Sun, 24 Nov 2002 11:42:48 -0500 (EST) · jop_at_astro.columbia.edu

>
> Concerning identifying the comp. stars used in future data submissions.  The
> different GSC catalogs use different numbering schemes, so would you prefer
> some sort of comp. star ID such as:
> GSC 2.2 N3320100520 so that you have a non-ambiguous identification?  Or is
> there a better (standard?) way to ID comp stars?

Thanks, Tom, for bringing this up.  Well for a long time I was totally
happy with "star 1.6' SSW of RU Loopy".  Then the GSC came along and
eventually it became routine to use it.  So many of our published papers
give the GSC name of our main comp.  Now we have GSC 2.2 with a bigger
name.  I don't object to this at all, but I see that observers are sending
data with the old numbering scheme.  So I don't want to plunge into any
new system until it's obvious that the average CBA observer can routinely
use it.  I just don't know that much about what kind of information
resources you have available at your desktop.  In fact I don't know that
much about what's available at *my* desktop, for that matter.

Anyway, I do want to stress that standardization of comparison stars is
not ultimately of much importance for our programs.  Why?  Because the
first thing I do to a long-night time series is to SUBTRACT THE MEAN AND
TREND.  This *greatly* improves the sensitivity to periodic signals in the
4-40 c/day range, usually our range of maximum interest.  Only for the
lowest frequencies does calibration matter.  And even for those, the best
calibration is achieved from simultaneous observation, even if the two
telescopes use different comparison stars.  I've noticed from experience
that the additive constants needed to splice data between telescopes vary
by 0.02-0.03 mag from night to night, and that this error is about the
same regardless of which comp star you use.  (The exact origin of it is
probably complex, but I suspect that the dominant element comes from the
very wide bandpass of unfiltered data - the price we pay to *have* a
powerful research program on CVs with small telescopes).

So the main message is just to *report* the comp star.  Since
Jonathan is archiving all data and we plan to make the time series
themselves retrievable, this is just good housekeeping.  The secondary
message is to try to use a common comp star, since that provides a crude
calibration even when no simultaneity is achieved (it's by no means always
achieved).  You can definitely use a different comp if you like, with no
significant loss in the science - but then stick to it.  As for the naming
scheme, I still like the GSC 1 or even the buried-treasure direction...
but I'll make it my business to figure out whatever newfangled name you've
supplied.

There are some comp-star recommendations on the CBA website for many
stars.  At the bottom of the "charts" page, there's a clickable "Earlier
CBA charts".  They're not complete, but we thought at some point (on
grounds of color and brightness) these stars might make good comps.
But of course you know your equipment (and saturation limits, and night
quality) best, and your judgment is probably the most reliable guide.

   joe