Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-13058

Inconsistent aperture corrections in W44 reprocessing of HSC RC1

    Details

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

      Description

      Tim Morton analysed the W44 HSC RC1 reprocessing (DM-12388) for LDM-503-2.

      He noted that object-level aperture corrections are not consistent:

      Gaussian - psfMag is whacked for all filters, as is Kron - psfMag. Worst near the edges. CModel looks to be OK.

      Investigate and figure out what's causing this issue.

        Attachments

        1. Screenshot 2017-12-08 09.07.38.png
          977 kB
          Lauren MacArthur
        2. Screenshot 2017-12-08 09.19.51.png
          840 kB
          Lauren MacArthur
        3. WIDE_8766_CModel_diff.png
          848 kB
          Lauren MacArthur
        4. WIDE_8766_CModel_sky_stars.png
          1.46 MB
          Lauren MacArthur
        5. WIDE_8766_forced_vs_unforced_CModel.png
          1.11 MB
          Lauren MacArthur
        6. WIDE_8766_forced_vs_unforced_Gaussian.png
          1.00 MB
          Lauren MacArthur
        7. WIDE_8766_Gauss_diff.png
          1.02 MB
          Lauren MacArthur
        8. WIDE_8766_Gaussian_sky_stars.png
          1.39 MB
          Lauren MacArthur

          Issue Links

            Activity

            Hide
            swinbank John Swinbank added a comment -

            Per discussion of 2017-12-18, this effect is seen only in forced measurement.

            Show
            swinbank John Swinbank added a comment - Per discussion of 2017-12-18, this effect is seen only in forced measurement.
            Hide
            price Paul Price added a comment -

            Lauren MacArthur notes:

            It looked to me that whatever is going wrong is happening only in the non-ref band (i.e. I band looked ok)

            And Jim Bosch pointed out:

            It also only affected algorithms that rely on the slot shape to set their effective aperture.

            Show
            price Paul Price added a comment - Lauren MacArthur notes: It looked to me that whatever is going wrong is happening only in the non-ref band (i.e. I band looked ok) And Jim Bosch pointed out: It also only affected algorithms that rely on the slot shape to set their effective aperture.
            Hide
            lauren Lauren MacArthur added a comment -

            I sort of take it back...in this plot:

            HSC-I looks much better than than the others (modulo the bright end jump), but there is a significant offset. And note that CModel looks ok in all bands:

            The above is for COSMOS. I will also attach a WIDE tract.

            Show
            lauren Lauren MacArthur added a comment - I sort of take it back...in this plot: HSC-I looks much better than than the others (modulo the bright end jump), but there is a significant offset. And note that CModel looks ok in all bands: The above is for COSMOS. I will also attach a WIDE tract.
            Hide
            lauren Lauren MacArthur added a comment -

            Here is a WIDE tract:
            Gaussian Mag - Psf Mag:

            CModel Mag - Psf Mag:

            Show
            lauren Lauren MacArthur added a comment - Here is a WIDE tract: Gaussian Mag - Psf Mag: CModel Mag - Psf Mag:
            Hide
            lauren Lauren MacArthur added a comment -

            And you can see the same pattern in the forced vs. unforced mags (hence the conclusion that this issue is only occurring in the forced measurements):

            Show
            lauren Lauren MacArthur added a comment - And you can see the same pattern in the forced vs. unforced mags (hence the conclusion that this issue is only occurring in the forced measurements):
            Hide
            jbosch Jim Bosch added a comment - - edited

            Robert Lupton and I discussed this today, and came to the following conclusions:

            • As several people have pointed out, we should expect there to be an offset between Kron and PSF and between Gaussian and PSF in the forced measurements. Both Kron and Gaussian have per-object apertures but do not account for the PSF, and the aperture corrections we're using do not work when that per-object aperture is calculated in a band (and hence with a PSF) other than the one we're photometering.
            • What we don't understand is why we can see multiple loci (apparently corresponding to different choices for the reference band) in just the population of bright stars. Only faint objects near the i-band detection limit should have some other band as their reference band.

            To debug, then, I think we want to identify some of these bright stars with non-i reference bands, look at images of them in all bands, inspect the reference-band-selection code, and generally figure out what's going wrong there.

            Show
            jbosch Jim Bosch added a comment - - edited Robert Lupton and I discussed this today, and came to the following conclusions: As several people have pointed out, we should expect there to be an offset between Kron and PSF and between Gaussian and PSF in the forced measurements. Both Kron and Gaussian have per-object apertures but do not account for the PSF, and the aperture corrections we're using do not work when that per-object aperture is calculated in a band (and hence with a PSF) other than the one we're photometering. What we don't understand is why we can see multiple loci (apparently corresponding to different choices for the reference band) in just the population of bright stars. Only faint objects near the i-band detection limit should have some other band as their reference band. To debug, then, I think we want to identify some of these bright stars with non-i reference bands, look at images of them in all bands, inspect the reference-band-selection code, and generally figure out what's going wrong there.
            Hide
            price Paul Price added a comment - - edited

            I believe the answer is in the stripy pattern of detection, which might be explained by saturation trails in the coadds.

            Show
            price Paul Price added a comment - - edited I believe the answer is in the stripy pattern of detection, which might be explained by saturation trails in the coadds.

              People

              • Assignee:
                Unassigned
                Reporter:
                swinbank John Swinbank
                Watchers:
                Jim Bosch, John Swinbank, Lauren MacArthur, Paul Price
              • Votes:
                0 Vote for this issue
                Watchers:
                4 Start watching this issue

                Dates

                • Created:
                  Updated:

                  Summary Panel