Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-23641

DRP Analysis Sprint Feb 2020: Quantify Local Sky Over-Subtraction

    XMLWordPrintable

Details

    • Story
    • Status: Done
    • Resolution: Done
    • None
    • None
    • None
    • 8
    • DRP S20-3 (Feb)
    • Data Release Production
    • No

    Description

      Can sky object statistics be used to help quantify how much flux is being over-/under-subtracted around bright sources?

      Here we use sky objects generated on DEEP data to explore the sky level ('local sky') as a function of regional source flux. A sky object is a pseudo-object artifically injected into the coadd processing run prior to the measurement step. They are defined to have a radius of 8 pixels, and are placed such that no pixel within this radius overlaps with any other detection footprint. Pointings are tested as to whether or not they meet this criteria, stopping when it reaches 100 successful pointings per patch or 500 attempts, whichever comes first.

      In the first half of this notebook, we create a re-run of DM-21386 for tract 9813, patch 2,2 in the r-band on the command line. The second half of this notebook expands our analyses to the entire tract.

      All outputs (including processing notebooks) can be found on GitHub.

      Attachments

        1. sky_object_flux_offset.png
          sky_object_flux_offset.png
          339 kB
        2. skyobject14.png
          skyobject14.png
          554 kB
        3. skyobject20.png
          skyobject20.png
          378 kB
        4. skyobject28.png
          skyobject28.png
          551 kB

        Issue Links

          Activity

            lskelvin Lee Kelvin added a comment -

            Thanks for reviewing this Jeff. The repo contains two notebooks: main and scrapbook. Scrapbook contains all of my working, additional plots, things I tried that didn't add to the final result, etc. I include it here, for reference, but the main notebook should be all you need.

            lskelvin Lee Kelvin added a comment - Thanks for reviewing this Jeff. The repo contains two notebooks: main and scrapbook. Scrapbook contains all of my working, additional plots, things I tried that didn't add to the final result, etc. I include it here, for reference, but the main notebook should be all you need.

            This notebook is fantastic - thorough, detailed, and clear. Certainly provides a lot of food for thought. Just a couple brief (non-essential) comments, and then it's good to go:
             
            "skyerr = 0.0022 # typical error on the sky for these data (ADU)" – Maybe add a comment on where this comes from?
             
            For the uninitiated, it also might be worth noting that the difference between the "deepCoadd" and "deepCoadd_calexp" is background subtraction (and that the sky objects should have, on average, zero flux in the calexp). 
             
            For reference, it may also be useful to note the aperture sizes in arcsec (maybe display a small table?).

            jcarlin Jeffrey Carlin added a comment - This notebook is fantastic - thorough, detailed, and clear. Certainly provides a lot of food for thought. Just a couple brief (non-essential) comments, and then it's good to go:   "skyerr = 0.0022 # typical error on the sky for these data (ADU)" – Maybe add a comment on where this comes from?   For the uninitiated, it also might be worth noting that the difference between the "deepCoadd" and "deepCoadd_calexp" is background subtraction (and that the sky objects should have, on average, zero flux in the calexp).    For reference, it may also be useful to note the aperture sizes in arcsec (maybe display a small table?).
            lskelvin Lee Kelvin added a comment -

            Thanks for the feedback Jeff. Re: the skyerr, I simply take the median 1-sigma standard deviation as derived for each of the bins in the magnitude range 18 < m < 19 (the working is in the scrapbook, line 70). A bit quick and dirty, but I couldn't figure out a better way on a short timescale. I'm open to suggestions if you have something else in mind?

            I'll work your suggestions in and merge it later today, cheers!

            lskelvin Lee Kelvin added a comment - Thanks for the feedback Jeff. Re: the skyerr, I simply take the median 1-sigma standard deviation as derived for each of the bins in the magnitude range 18 < m < 19 (the working is in the scrapbook, line 70). A bit quick and dirty, but I couldn't figure out a better way on a short timescale. I'm open to suggestions if you have something else in mind? I'll work your suggestions in and merge it later today, cheers!

            I don't think there's a problem with the sky error - just thought it's worth noting how you derived it.

            jcarlin Jeffrey Carlin added a comment - I don't think there's a problem with the sky error - just thought it's worth noting how you derived it.
            lskelvin Lee Kelvin added a comment -

            Thanks again. I've added extra info about the items you mention into the main notebook, merged to master, and deleted the branch. Cheers!

            lskelvin Lee Kelvin added a comment - Thanks again. I've added extra info about the items you mention into the main notebook, merged to master, and deleted the branch. Cheers!

            People

              lskelvin Lee Kelvin
              lskelvin Lee Kelvin
              Jeffrey Carlin
              Jeffrey Carlin, Lee Kelvin
              Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved:

                Jenkins

                  No builds found.