The NOAO Deep Wide-Field Survey MOSAIC Data Reductions

by Buell T. Jannuzi, Jenna Claver, and Frank Valdes

Version 7.02 (June 2003) -- This is the current version of the NOAO Deep Wide-Field Survey (NDWFS) MOSAIC reduction notes. For anyone actively working on a data set using version 6.2 (October 9, 2000) -- which is now out of date -- it is still available on the web, but I expect to remove it from the Web at the end of 2002. Send comments/corrections for Version 7 to Buell Jannuzi at

IMPORTANT The biggest differences between version 7.0 and 6.2 of this guide result from the upgrades to IRAF (to version 2.12) and mscred (to version 4.7) detailed further below and at this link from Frank Valdes, New Mosaic Reduction Capabilities, but other steps have changed as well. In particular, many steps are less interactive and make use of IRAF's ability to handle masks to improve the quality of the reductions. We expect most users of MOSAIC will want to revise their reduction procedures to make use of these improvements.


NOAO's two MOSAIC cameras produce very large images at a significant rate, resulting in a large, but manageable, data reduction challenge. In these Web pages we describe the current proceedures for the reduction and calibration of the NOAO Deep Wide-field Survey (NDWFS) Mosaic images. These notes were generated by Buell Jannuzi, Jenna Claver, and Frank Valdes based on what the NDWFS has learned from reducing MOSAIC images using the IRAF tasks written by Frank Valdes, Lindsey Davis, Doug Tody, and the rest of the IRAF group. Additional scripts by Ed Ajhar, Ian Dell'Antonio, Arjun Dey, Marc Dickinson, Chris Greer, Buell Jannuzi, and James Rhoads have been used during various stages of our development of these reduction procedures, although the specific scripts have generally now been replaced by IRAF tasks. An earlier version of this guide was written by Buell Jannuzi and Chris Greer (then an REU summer student at KPNO), and this version makes use of portions of that guide. Additional input on techniques/methods/steps in the reduction path were contributed or adopted by BTJ from the work of Ed Ajhar, Tod Lauer, and Phil Massey. Very helpful information about the behavior of the two MOSAIC cameras has also been provided by Taft Armandroff, Chris Smith, George Jacoby, Rich Reed, Tom Wolfe, Roger Smith, Steve Heathcote, and Alistair Walker. Users of earlier versions of these notes (including Michael Brown, Heather Groch, Dan Macintosh, Phil Massey, Daniel Stern, Aaron LaCluyze, John Silverman and Steve Dawson) have helped identify sections that needed improvement -- some of these have been addressed in this version of the guide. Hopefully these notes will continue to improve as we all learn to improve our reductions of the MOSAIC images.

These notes are not a replacement for the help that can be found in the notes by Frank Valdes about the mscred reduction package in IRAF. Frank's notes (not all yet current, but still useful) can be found by typing mscguide at the cl> prompt in IRAF or by this link here. Note that there are versions of the new mscred package that work under V2.11.3 as well as V2.12 -- but that not all of the new features will be available under V2.11.3. Additional documentation about IRAF for the Mosaic cameras, in particular the article by Valdes, "The Reduction of CCD Mosaic Data", can be found here. Further, these pages do not attempt to justify why we apply the basic reduction steps discussed here. For such discussions see the excellent guide by Phil Massey, A User's Guide to CCD Reductions with IRAF and the NOAO CCD Mosaic Imager User Manual . These notes can be used as one practical example of using the available tools in IRAF to apply basic calibrations to a set of MOSAIC images. They are not intended to be an exhaustive listing of all necessary reductions you might have to make for your own particular science application.

Since the first version of our MOSAIC reduction guide, other groups have used it and then set up their own guides to detail how they have decided to modify our approahes to meet their specific goals (or because they disagree with one or more of our choices, why they developed new approaches). Examples (not exhaustive by any means) of such guides can be found at these links: Gilles Bergond's Reducing CCD Mosaic Images: A Beginner's Cookbook, the Local Group Survey Mosaic Reduction Notes , and notes found off the Deep Lens Survey web page.

Note that occasionally you will see links starting with NDWFS. These point to NOAO Deep Wide-Field Survey pages that are very specific to our survey and unlikely to be of general interest.

Last Modified June 24, 2003 by Jannuzi -- corrected a significant typo in the formula for weighting the individual images combined into the final image stack for a field. Thank you to Eric Gawiser for pointing this out.