Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-17536

Investigate flux correlations in HiTS processing

    Details

    • Story Points:
      4
    • Epic Link:
    • Sprint:
      AP S19-3, AP S19-4, AP S19-5
    • Team:
      Alert Production

      Description

      As part of DM-16406 Meredith Rawls identified systematic correlations between the lightcurves of various HiTS sources. This ticket is to investigate the cause of that correlation.

        Attachments

          Issue Links

            Activity

            Hide
            mrawls Meredith Rawls added a comment -

            I plotted a handful of DIA Object light curves from the HiTS 2015 dataset (see: correlated_lightcurves.png), and found that indeed there is a troubling correlation between them. After discussions with Chris Morrison and Eric Bellm, I determined the problem seems to stem from the calib object having a bad flux zeropoint (see: bad_flux_zeropoints.png). It is worth noting that the calib information is not saved in the PPDB and must be loaded from each exposure instead. Additional work needs to be done to investigate what is causing the poor photometric calibration (DM-18687).

            Show
            mrawls Meredith Rawls added a comment - I plotted a handful of DIA Object light curves from the HiTS 2015 dataset (see: correlated_lightcurves.png), and found that indeed there is a troubling correlation between them. After discussions with Chris Morrison and Eric Bellm , I determined the problem seems to stem from the calib object having a bad flux zeropoint (see: bad_flux_zeropoints.png). It is worth noting that the calib information is not saved in the PPDB and must be loaded from each exposure instead. Additional work needs to be done to investigate what is causing the poor photometric calibration ( DM-18687 ).
            Hide
            ebellm Eric Bellm added a comment -

            I will mark this ticket as closed but append some clarifications to the explanation--namely that the flux correlations are seen in the totFlux measurements (forced measurements on the calexps) in addition to the psfFlux measurements on the differences, indicating the problem lies in the calexps and not in the diffim; and that varying zeropoints are not themselves indicative of bad photometric calibration (maybe there were clouds!).

            Show
            ebellm Eric Bellm added a comment - I will mark this ticket as closed but append some clarifications to the explanation--namely that the flux correlations are seen in the totFlux measurements (forced measurements on the calexps) in addition to the psfFlux measurements on the differences, indicating the problem lies in the calexps and not in the diffim; and that varying zeropoints are not themselves indicative of bad photometric calibration (maybe there were clouds!).
            Hide
            mrawls Meredith Rawls added a comment -

            Thanks for this important clarification!

            Show
            mrawls Meredith Rawls added a comment - Thanks for this important clarification!

              People

              • Assignee:
                mrawls Meredith Rawls
                Reporter:
                ebellm Eric Bellm
                Reviewers:
                Eric Bellm
                Watchers:
                Eric Bellm, Meredith Rawls
              • Votes:
                0 Vote for this issue
                Watchers:
                2 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved: