Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-3417

Fix overestimation of aperture correction error

    Details

    • Type: Story
    • Status: To Do
    • Resolution: Unresolved
    • Fix Version/s: None
    • Component/s: meas_base
    • Labels:
      None

      Description

      We're overestimating the aperture correction errors by including photon noise twice: both in the aperture correction errors and the original measurement errors.

      This is a migration of HSC-1277 to LSST, which is where it should be fixed.

        Attachments

          Issue Links

            Activity

            Hide
            yusra Yusra AlSayyad added a comment -

            This will be a multi-ticket project. 

            Show
            yusra Yusra AlSayyad added a comment - This will be a multi-ticket project. 
            Hide
            swinbank John Swinbank added a comment -

            Michael Wood-Vasey – Robert suggests that we should let you know that this issue exists!

            Show
            swinbank John Swinbank added a comment - Michael Wood-Vasey – Robert suggests that we should let you know that this issue exists!
            Hide
            swinbank John Swinbank added a comment -

            On the original HSC ticket, JimB writes:

            After a lot of mathematics and discussion with Robert Lupton, I'm pretty sure that the formula in the current code assumes:

            • the corrected and measured flux are both aperture fluxes OR the noise from the background is more important than the noise from the source
            • the actual aperture correction is close to 1

            I'm a little worried we're not that close to these conditions (especially the first one, since we're measuring the aperture corrections on bright sources). It's pretty easy to fix, if I compute one more quantity in a couple of the flux measurement algorithms.

            Show
            swinbank John Swinbank added a comment - On the original HSC ticket, JimB writes: After a lot of mathematics and discussion with Robert Lupton, I'm pretty sure that the formula in the current code assumes: the corrected and measured flux are both aperture fluxes OR the noise from the background is more important than the noise from the source the actual aperture correction is close to 1 I'm a little worried we're not that close to these conditions (especially the first one, since we're measuring the aperture corrections on bright sources). It's pretty easy to fix, if I compute one more quantity in a couple of the flux measurement algorithms.

              People

              • Assignee:
                jbosch Jim Bosch
                Reporter:
                swinbank John Swinbank
                Watchers:
                John Swinbank, Michael Wood-Vasey, Yusra AlSayyad
              • Votes:
                0 Vote for this issue
                Watchers:
                3 Start watching this issue

                Dates

                • Created:
                  Updated:

                  Summary Panel