Re: Data reduction methodology for V-I colors

[Thomas Droege <tdroege2@earthlink.net>]

Mike,

Yep, this and a lot more.  I cannot identify a "bad" night.  I did a lot of
work on the assumption that there was such a thing as a bad night.  For
example, go through all the data and look for stars with big deviations
from the mean.  (This should work because most stars are not variable.) Now
sort all frames by fraction of stars with big deviations from the mean. 
Now eliminate these from the data set and look at the result. I did this
eliminating the noisiest 10%, 20%, 50% ...  No improvement at any cut
level.  Conclusion:  Eliminating frames with lots of deviant measurements
does not improve the quality of the data as a whole. No doubt eliminating
noisy frames will improve the data when we solve other problems.  But noisy
frames are not even close to being the largest source of error or the above
experiment would have made some improvement.  It made no improvement.   I
was surprised.  But when you think about it, the star measurement scheme
that Michael uses really does a good job.  It is something else.  Something
fundamental.  Something related to position in the frame.  If you track a
field, the result is much better.  But this just hides the problem.  

I think there are two likely suspects.  I am working on one, Michael is
working on the other.

1) There is some gradient in the sky that produces errors in all frames.

Michael is working on this.

2) We don't have a good reference catalog.  The result is that depending on
where a star is in a frame, a different set of reference stars are used. 
If the mean of the reference star set is different for the two frame
positions, then this will produce an error.

I am working on this.

OK, I suspect that both have an effect and that there are still more
problems to be found.  I know that I have made some small improvements
using 2.  I am just waiting for more data and winter time with no
observations to sit and compute on this for a couple of months.

Tom Droege




> [Original Message]
> From: Michael Koppelman <lolife@bitstream.net>
> To: Tass <tass@listserv.wwa.com>
> Date: 8/31/2004 11:57:14 AM
> Subject: Re: Data reduction methodology for V-I colors
>
> This got me thinking and I'm sure there are tech notes and such I could 
> find about it but nonetheless -- have you guys considered calibrating 
> the night rather than the frame? i.e. take the whole dataset, read in 
> all the stars from the entire night, throw out the ones with high 
> errors, match up the Landolt/Henden or other standards and get 
> transforms for the whole night?
>
> Cheers,
> Michael
>