stscijgbot-hstdp
commented
2 years ago Comment by Warren Hack on JIRA:
Fundamentally, the problem with processing 'ie3501buq' stems from problems with the a posteriori WCS solution for alignment to GAIAeDR3 provided by the astrometry database. That specific headerlet in the database is missing a lot of information for a proper WCS solution; namely, a non-SIP solution is provided, whereas the same solution for the rest of the exposures all have a complete SIP solution. A complete SIP solution includes all the SIP distortion keywords (A{}{}{}, B{}{*}{*}) along with a value for CTYPE of 'RA-TAN—SIP'/'DEC-TAN–-SIP'. The headerlet from the database for 'ie3501buq' only has:
XTENSION= 'IMAGE ' / Image extension BITPIX = 8 / array data type NAXIS = 0 / number of array dimensions PCOUNT = 0 / number of parameters GCOUNT = 1 / number of groups WCSAXES = 2 / Number of coordinate axes CRPIX1 = 507.0 / Pixel coordinate of reference point CRPIX2 = 507.0 / Pixel coordinate of reference point CUNIT1 = 'deg' / Units of coordinate increment and value CUNIT2 = 'deg' / Units of coordinate increment and value CTYPE1 = 'RA---TAN' / Right ascension, gnomonic projection CTYPE2 = 'DEC--TAN' / Declination, gnomonic projection CRVAL1 = 10.710735026267 / [deg] Coordinate value at reference point CRVAL2 = 40.888019404297 / [deg] Coordinate value at reference point CRDER1 = 58.843753099828 / [deg] Random error in coordinate CRDER2 = 64.378193953688 / [deg] Random error in coordinate WCSNAME = 'IDC_w3m18525i-FIT_REL_GAIAeDR3' / Coordinate system title MJDREF = 0.0 / [d] MJD of fiducial time RADESYS = 'ICRS' / Equatorial coordinate system ORIENTAT= -66.11076802765093 / positions angle of image y axis (deg. e of n) TG_ENAME= 'SCI ' / Target science data extname TG_EVER = 1 / Target science data extver CD1_1 = -1.5315828528647E-05 / partial of first axis coordinate w.r.t. x CD1_2 = -3.0711151841693E-05 / partial of first axis coordinate w.r.t. y CD2_1 = -3.4363220457108E-05 / partial of second axis coordinate w.r.t. x CD2_2 = 1.360240460729E-05 / partial of second axis coordinate w.r.t. y OCX10 = 0.000777297012973576 / original linear term from IDCTAB OCX11 = 0.1354298144578934 / original linear term from IDCTAB OCY10 = 0.1209627389907837 / original linear term from IDCTAB OCY11 = -0.00042068894254043 / original linear term from IDCTAB IDCSCALE= 0.1282500028610229 / pixel scale from the IDCTAB reference file EXTNAME = 'SIPWCS ' / extension name EXTVER = 1 / extension value
In comparison, it should contain the same set of keywords found in the GAIAeDR3 headerlet for 'ie3501c0q' (from the same ASN):
XTENSION= 'IMAGE ' / Image extension BITPIX = 8 / array data type NAXIS = 0 / number of array dimensions PCOUNT = 0 / number of parameters GCOUNT = 1 / number of groups WCSAXES = 2 / Number of coordinate axes CRPIX1 = 507.0 / Pixel coordinate of reference point CRPIX2 = 507.0 / Pixel coordinate of reference point CUNIT1 = 'deg' / Units of coordinate increment and value CUNIT2 = 'deg' / Units of coordinate increment and value CTYPE1 = 'RA---TAN-SIP' / TAN (gnomonic) projection + SIP distortions CTYPE2 = 'DEC--TAN-SIP' / TAN (gnomonic) projection + SIP distortions CRVAL1 = 10.710716583382 / [deg] Coordinate value at reference point CRVAL2 = 40.887792399172 / [deg] Coordinate value at reference point CRDER1 = 58.843753099828 / [deg] Random error in coordinate CRDER2 = 64.378193953688 / [deg] Random error in coordinate WCSNAME = 'IDC_w3m18525i-FIT_REL_GAIAeDR3' / Coordinate system title MJDREF = 0.0 / [d] MJD of fiducial time RADESYS = 'ICRS' / Equatorial coordinate system A_ORDER = 4 / SIP polynomial order, axis 0, detector to sky A_0_2 = 5.03683559395408E-08 / SIP distortion coefficient A_0_3 = 3.58489019323397E-11 / SIP distortion coefficient A_0_4 = -1.7096030015339E-13 / SIP distortion coefficient A_1_1 = 2.44067589697849E-05 / SIP distortion coefficient A_1_2 = 2.52491804653495E-11 / SIP distortion coefficient A_1_3 = 5.60344668392659E-13 / SIP distortion coefficient A_2_0 = -2.0586281536303E-07 / SIP distortion coefficient A_2_1 = 1.11756131621297E-10 / SIP distortion coefficient A_2_2 = 9.44397763409537E-15 / SIP distortion coefficient A_3_0 = -2.1153064583223E-10 / SIP distortion coefficient A_3_1 = 5.44164224825977E-13 / SIP distortion coefficient A_4_0 = -2.6241913593392E-13 / SIP distortion coefficient B_ORDER = 4 / SIP polynomial order, axis 1, detector to sky B_0_2 = 2.99526910047536E-05 / SIP distortion coefficient B_0_3 = -2.276184347652E-10 / SIP distortion coefficient B_0_4 = 7.38617646709978E-13 / SIP distortion coefficient B_1_1 = -1.8441031369285E-07 / SIP distortion coefficient B_1_2 = 5.41334217303662E-11 / SIP distortion coefficient B_1_3 = -2.2228484499494E-14 / SIP distortion coefficient B_2_0 = 6.96937375511744E-06 / SIP distortion coefficient B_2_1 = -3.0390505421181E-10 / SIP distortion coefficient B_2_2 = -2.0599022983307E-13 / SIP distortion coefficient B_3_0 = 3.38549135278375E-11 / SIP distortion coefficient B_3_1 = 1.14334174473825E-13 / SIP distortion coefficient B_4_0 = -6.4924382260858E-13 / SIP distortion coefficient ORIENTAT= -66.11564798546388 / positions angle of image y axis (deg. e of n) TG_ENAME= 'SCI ' / Target science data extname TG_EVER = 1 / Target science data extver CD1_1 = -1.5316430770876E-05 / partial of first axis coordinate w.r.t. x CD1_2 = -3.0706450875166E-05 / partial of first axis coordinate w.r.t. y CD2_1 = -3.4355129371857E-05 / partial of second axis coordinate w.r.t. x CD2_2 = 1.3597194238271E-05 / partial of second axis coordinate w.r.t. y OCX10 = 0.000777297012973576 / original linear term from IDCTAB OCX11 = 0.1354298144578934 / original linear term from IDCTAB OCY10 = 0.1209627389907837 / original linear term from IDCTAB OCY11 = -0.00042068894254043 / original linear term from IDCTAB IDCSCALE= 0.1282500028610229 / pixel scale from the IDCTAB reference file EXTNAME = 'SIPWCS ' / extension name EXTVER = 1 / extension value
This ultimately needs to be fixed in the astrometry database. It is not clear what else can be done within 'runastrodriz' pipeline processing to generate a better result. One option would be to update the header rules for fitsblender to not even have rule for the CTYPE keywords and rely on the processing code to set the WCS keywords entirely. This option would be implemented under HLA-769, where it will be tested to see whether or not it solves this problem.
Issue HLA-772 was created on JIRA by Michele De La Pena:
This visit is comprised of 12 images (4 images for each of three filters). The images are part of three associations and one singleton (ie3501010, ie3501020, ie3501030, and ie3501cdq). Image ie3501buq_flt.fits does not have the same CTYPE1/CTYPE2 entries as all of the other images after the astrodrizzle step. Why? See Jira HLA-760 for more information.