Re: GSC 1040-399 + new variable

["Thomas Droege" <tdroege2@earthlink.net>]

> You get that automatically, don't you?  If the star is not variable,
> then by definition it is constant.

I think not quite.  The; problem is that when you take data every night,
all the data is variable.  So one needs a way to sort out the "noise" from
the data.  That is what I am working on at the moment.  Since there are
several things going on at once, no one project "cleans" the data.  Michael
is working on the photometry.  Hopefully he will help make data taken on
good nights consistent.  I have been mostly working on "bad" data nights. 
Clouds, ice crystals, bad shutters, and the like.  

There is already a first cut in the pipeline thanks to Michael.  Setting
the parameters for this cut is not so easy.  A bad choice means thousands
of hours of work to try another.   I have set them, and I hope I have made
a good choice.  If you set them too tight, then data is lost that might be
useful.  If you set them too loose, then there is a clean up process that
is needed.  That is what I am working on.

Tom Droege

> [Original Message]
> From: Arne Henden <aah@nofs.navy.mil>
> To: <tdroege2@earthlink.net>
> Cc: Michael Koppelman <lolife@bitstream.net>; tass <tass@listserv.wwa.com>
> Date: 7/16/2003 11:55:14 AM
> Subject: Re: GSC 1040-399 + new variable
>
> Thomas Droege wrote:
> > One of the uses of the tass data is to find good comparison stars. 
Several
> > of us are working to filter the tass data to find the stars that do not
> > seem to vary.  In a way this might be as useful as the variable data. 
;^)
> > 
>
> You get that automatically, don't you?  If the star is not variable,
> then by definition it is constant.
>    As I have often cautioned people using my 1-2-3 night sequences, they
> do not guard against variability of any particular star.  For the
> Henden/Honeycutt sequence papers, we used Kent's Roboscope multiyear
> dataset to check for variability and deleted any possible variables
> from the sequences.
>    Tom's multiyear (plus, of course, other Mark IV sites) datasets will
> be useful in looking for the really long-term variations, such as
> starspot activity like the solar cycle, or finding those year-plus
> eclipsing systems.  The ROTSE dataset (when/if it becomes available)
> will do the same.  The long-term variables (not Miras!) are an
> uncharted regime in the all-sky sense (MACHO and OGLE amongst others
> have multiyear datasets in small regions of the sky).
>    The real benefit IMHO of survey work is not the common variable
> like EBs, Miras or RR Lyrs, but the unusual star.  With many millions
> of potential candidates, you can afford to ignore 99.999percent of
> all stars and concentrate on a handful of interesting ones.  Let
> someone else study the common stars; skim the cream.
> Arne
>