Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-19978

Fix bug in DcrModel coordinates

    Details

    • Story Points:
      4
    • Epic Link:
    • Sprint:
      AP F19-1
    • Team:
      Alert Production

      Description

      A bug in the calculation of the direction of DCR resulted in the algorithm failing to converge for some observations. The fix is to write a new unit test and expand existing tests to cover more of the coordinate parameter space, and make changes as necessary to pass the tests.

        Attachments

          Issue Links

            Activity

            Hide
            sullivan Ian Sullivan added a comment -

            This fixes a bug that caused the direction of DCR to be calculated incorrectly for Decam data. It also expands the unit tests to make sure they would catch this and similar bugs in the future.

            However, I was unable to rely solely on astropy coordinate transformations, since the calculations for hour angle and LST fail the simple testDummyVisitInfo unit test by several arcseconds, even after correcting for precession and nutation. Perhaps I am missing something, so I would welcome suggestions on how to better construct the dummy visitInfo without my hour angle calculation.

            Show
            sullivan Ian Sullivan added a comment - This fixes a bug that caused the direction of DCR to be calculated incorrectly for Decam data. It also expands the unit tests to make sure they would catch this and similar bugs in the future. However, I was unable to rely solely on astropy coordinate transformations, since the calculations for hour angle and LST fail the simple testDummyVisitInfo unit test by several arcseconds, even after correcting for precession and nutation. Perhaps I am missing something, so I would welcome suggestions on how to better construct the dummy visitInfo without my hour angle calculation.
            Hide
            sullivan Ian Sullivan added a comment -

            We determined that the calculations of hour angle and parallactic angle in lsst.afw.image.VisitInfo.cc are incorrect, because it uses ICRS RA and Dec coordinates. Since this ignores precession it results in an error of up to 10 arcminutes in parallactic angle for 2020 observations, but as demonstrated by the new unit tests written for this ticket the error in the DCR shift calculation is negligible. We will address the incorrect calculations in DM-20133.

            Show
            sullivan Ian Sullivan added a comment - We determined that the calculations of hour angle and parallactic angle in lsst.afw.image.VisitInfo.cc are incorrect, because it uses ICRS RA and Dec coordinates. Since this ignores precession it results in an error of up to 10 arcminutes in parallactic angle for 2020 observations, but as demonstrated by the new unit tests written for this ticket the error in the DCR shift calculation is negligible. We will address the incorrect calculations in DM-20133 .

              People

              • Assignee:
                sullivan Ian Sullivan
                Reporter:
                sullivan Ian Sullivan
                Reviewers:
                Tim Jenness
                Watchers:
                Ian Sullivan, John Swinbank, Tim Jenness
              • Votes:
                0 Vote for this issue
                Watchers:
                3 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved:

                  Summary Panel