Data Reduction Frequently Asked Questions

I see a systematic offset in the coordinates determined for my extracted spectra with respect to the known coordinates of my objects - why?


This shift can be due to 3 effects:

  1. The limited accuracy of the VLT pointing and guide star coordinates. This can on average be about 2-3 arcsec and at maximum about 5 arcsec.
  2. The pipeline uses the undeviated wavelength of the grism (stored in the keyword HIERARCH ESO INS GRIS1 WLEN) to determine the position of the object for a given extracted spectrum. We found that the original values are not accurate for some grisms. For data taken since April 5, 2017, the header values should be correct. The older data are being updated but due to the data volume this will take some time. The new values are listed below. If you change the header values to the new ones your WCS should improve.
    GRIS_1200B +97 free 436.  
    GRIS_1200g +96 free 486.  
    GRIS_1200R +93 GG435 653.  
    GRIS_1400V +18 free 520.3  
    GRIS_150I +27 [all] 716.9
    GRIS_200I +28 free 743.2  
    GRIS_300V +20 [all] 585.4 until 2009-03-03
    GRIS_300V +10 [all] 584.9 since 2009-03-04
    GRIS_300I +21 [all] 857.3 until 2009-03-03
    GRIS_300I +11 [all] 857.5 since 2009-03-04
    GRIS_600B +22
    free 462.7  
    GRIS_600V +94 GG435 581.  
    GRIS_600R +14 GG435 627.7  
    GRIS_600RI +19 GG435 677.7  
    GRIS_600I +25 OG590 793.5  
    GRIS_600z +23 OG590 892.7
  4. The wavelength calibration is not accurate enough. As stated above the pipeline uses the wavelength recorded in HIERARCH ESO INS GRIS1 WLEN to calculate the coordinates for an extracted spectrum. If the wavelength calibration is not good this will introduce an offset.