DIS III - Data Reduction Guide

Contact Webmaster for web page errors, corrections, ommisions, and/or additions.

This page last updated: March 3, 2008 - JMD
This page last checked: September 27, 2004 - JMD

Printer Friendly version of this page

This page is no longer being updated.

Please refer to DIS Users Guide

Contents

Data Reduction Guide

Gordon Richard's Reduction Guide (Outdated)

Data Reduction Guide

OUTDATED!!!

Now for an extra special treat I will give you the accepted parameters for the overscan regions (to be found nowhere else as far as I can tell!) For the RED chip the overscan region is [855:870,*] and for the BLUE chip it is [570:585,*].

Converting FITS Images to IRAF Images

The first thing to do is change the .fits images into IRAF .imh images. To do this create files (in the directory where all your images are) that has the names of all of the different types of images you have (flat,bias,dark,raw) For example, for the RED images, files called, say, rbiasfitslist, rflatfitslist, and rimagefitslist might have the following lines in them:

disbiasr0001.fit
disbiasr0002.fits
disbiasr0003.fits

disflatr0001.fits
disflatr0002.fits
disflatr0003.fits

disimr0001.fits
disimr0002.fits
disimr0003.fits

disbiasr0001
disbiasr0002
disbiasr0003

disflatr0001
disflatr0002
disflatr0003

disimr0001
disimr0002
disimr0003

Now run IRAF by typing cl in the IRAF directory. Then change directories to the directory where all of your data resides. Load the data input-output package by typing dataio. The commands
rfits @rbiasfitslist * @rdisbiaslist
rfits @rflatfitslist * @rdisflatlist
rfits @rimagefitslist * @rdisimagelist
will convert all of the images to IRAF format. N.B. if disk space is a problem, you may want to delete the .fits images after you are done (only if they are on tape!), and you may want to delete the .fits images after each rfits command.

Making a Super-bias Frame

The next step is to make a "super-bias" frame which will be subtracted off of all the other frames. At this point you need to load the ccdredd package. Do this by typing
noao
imred
ccdred
The command
zerocombine @rdisbiaslist output=rbias combine="median" reject="sigclip" ccdtype="" clobber+ statsec=[###:###,###:###]
will produce a super-bias frame called "rbias.imh". Where ### should be filled in with the proper numbers for the good data section of the CCD.

De-biasing Images

With this super-bias frame you can de-bias all of your other images. To do this you'll have to run ccdproc. The commands that you want to use are
ccdproc @rdisflatlist ccdtype="" fixpix- overscan+ trim- zerocor+ darkcor- flatcor- biassec=[###:###,*] trimsec= [###:###,###:###] zero=rbias interactive-
ccdproc @rdisimagelist ccdtype="" fixpix- overscan+ trim- zerocor+ darkcor- flatcor- biassec=[###:###,*] trimsec= [###:###,###:###] zero=rbias interactive-

Making a Super-flat Frame

The next step is to create a "super-flat" with flatcombine. Here the command is
flatcombine @rdisflatlist output=rflat combine="median" reject="sigclip" ccdtype="" process- subsets- clobber+ scale="mode" statsec=[###:###,###:###]

Flat Fielding

Now that we have a super-flat, we want to use it to remove the pixel-to-pixel variation in the chip. We run ccdproc again to do this with the command
ccdproc @rdisimagelist ccdtype="" fixpix- overscan+ trim+ zerocor+ darkcor- flatcor+ biassec=[###:###,*] trimsec=[###:###,###:###] flat=rflat
All other parameters should be the same as the last pass.

Doing it All

One can actually put all of these steps into a single script. For example I have put all of the above lines into a file called cleanred.cl. Then to run this program simply go into the IRAF directory, and type
cl
dataio
noao
imred
ccdred
then change directories to where your data is, and type
cl < cleanred.cl

Now all of the images listed in the "rdisimagelist" file should all be properly cleaned-up. I should mention that this process suffices for Adler's needs and for a first cut on data reductions, but one should really do a more careful reduction.

Incidentally, there is no reason that this can't be used to clean up DSC and GRIM images. Just rename the files appropriately to avoid confusion, and remember that the DSC data section is [21:1044,1:1024] and for GRIM it is [###:###,###:###].

Oh, and the read-noise for both chips is 9.5 electrons. The gain on the BLUE chip is approximately 1.0 electrons/ADU, and on the RED chip is 1.83 electrons/ADU. In addition, the max counts you want in an exposure is 45,000 for the RED chip, and 55,000 for the BLUE chip.

14 May 1996

Please note that Ledlow, Holtzman and Watson give the RN for the blue chip as 16 electrons, and the gain as .99 electrons/ADU. For the read chip they give a RN of 15 electrons, and a gain of 1.98 electrons/ADU.