RE: Making Flats

["Michael Gutzwiller" <deepsky@fuse.net>]

One thing to keep in mind when making median flats.  I have found that
before you median combine the images you need to force the images to the
same level.  In other words there needs to be a step between Michael's steps
b and c.  Step b2 would be - scale each image so the median pixel is equal
to 10000 (or whatever number seems appropriate).

The reason you need to do this is to account for differing levels of
background sky for different times of the night.

Mike G.


-----Original Message-----
From: owner-tass@listserv.wwa.com [mailto:owner-tass@listserv.wwa.com]On
Behalf Of Stupendous Man
Sent: Monday, April 30, 2001 10:02 AM
To: tass@listserv.wwa.com
Cc: mwrsps@rit.edu; tdroege@veriomail.com
Subject: Re: Making Flats



  Tom had some questions about making flatfield frames
for disk set 18:

> 1) I then want to say to it, look on this disk, get all the object files,
> and make a median flat from them.  Then write it to this file.  Do this
> separately for the V and I files.

> 2) Look at the disk, find all the dark files and make a median dark from
> them.  Do this separately for the V and I files.

  I think the process is a little more complicated than that.
In order to make a flatfield image from a set of pictures of
the sky, one must

    a. combine a set of dark frames to create a "master dark"

    b. subtract the master dark from each of the sky images

    c. combine the sky images to create a "master flat"

  And, as Arne would point out, the flatfield image created in
this manner is not the best one can do.  The signal in each of
the sky frames is likely to be low, so that signal-to-noise in
each pixel of the median flatfield frame won't be very high.
In order to achieve the best results, one could fit some low-order
polynomial to this median sky flat, and combine it with a dome
flat of high signal-to-noise.  But that's a complicated business,
and I don't have time to get into it right now.

  So, let's continue with an example of making a simple median
sky flat.  Suppose that one has files on disk with names like:

           dark1.fit, dark2.fit, dark3.fit, ... , darkN.fit

           sky1.fit, sky2.fit, sky3.fit, ... , skyM.fit

  In the XVista package, one could perform the required steps
like so (where the ellipsis indicates a bunch of file names
that one must type, and comments to the reader are prefaced
by pound signs):

    # This creates the master dark frame
    median dark1.fit dark2.fit dark3.fit ... darkN.fit
outfile=master_dark.fit

    # This subtracts the dark from each of the sky frames
    #   (in this step, I assume that one is running the Bash shell)
    for i in sky*.fit
      do
      sub $i master_dark.fit
      done

    # This creates the median sky flatfield image
    median sky1.fit sky2.fit sky3.fit ... skyM.fit outfile=master_flat.fit


  In IRAF, one would use the "imcombine" task for the first and third
steps, and the "imarith" task for the second step.  IRAF has its own
way of performing a loop over a set of files, so the second step would
look quite different.

                                          Michael Richmond