Cut Settings

[Thomas Droege <tdroege2@earthlink.net>]

The pipeline keeps throwing out data that looks pretty good.  I have been
using the following settings for sky:

V level 4000  V sigma 120
I level 9000  I sigma 180

The I sigma cut at 180 seems to be throwing out images that look pretty
good.  These were arbitrary settings, picked from winter sky readings.  I
now think that this setting is a bit too tight.  I have changed it  to 200
from 180.

When analyzing last night's data, this produced 32,723 stars instead of the
16,909 stars measured with this cut at 180.

OK, this is an arbitrary biz.  One cannot determine that such;chi a
decision is correct without analyzing a ton of data.  I will run with this
setting for a while, and see if I can tell any difference.  I doubt it.  So
far almost nothing that I do makes any obvious difference in the data.  We
need a "quality meter" for the data.  What would be nice would be to be
able to take a single image and determine it's quality.  I have been having
a go at this. 

What I use to determine the overall quality is a mag vs mag sigma plot.  If
anyone has a better tool, let me know.  One does not see much in such a
plot without a lot of data.  Further, I have found little that I can do
that affects such a plot.  The ;main thing that improves it is to measure
the stars in the same position of the frame every time.  This produces
better relative photometry, but I think does nothing for the absolute
photometry.  I did not design the Mark IVs for accurate positioning to a
sky grid, so it is awkward for me to try to run in this mode.  My opinion
is that taking data in this way produces data that looks more accurate than
it really is.  OK, one can put caveats on the data, but; people will look
at the spread of the data points and assume accuracy that is not there.  

All this is just rationalization for why I want to run the way I am
running.  That is just scan what comes over stepping in declination and
without ;regard for the particular RA.

Tom Droege