I Release Notes—V3.1-1

The main change is the addition of support for IRIS2.

I.1 New recipes

ADDWCS

Adds a valid WCS in the FITS headers of raw data. This recipe is for all instruments.

ARRAY_TESTS

This is for IRIS2. It calculates and reports the readnoise and dark current using a group of four frames taken in the array tests sequence. The readnoise and dark current are logged to a text file, and the readnoise is filed with the calibration system.

DARK_SUBTRACT

Subtracts a dark frame from an observation. This recipe is meant for a quick look of data that are being taken at a high rate.

NOD_CHOP_FAINT

Reduction of nodded and chopped data, specifically for Michelle. It is similar to NOD_CHOP, but first removes horizontal and vertical artifacts from the mosaic, then forms a single image of the source using a median filter, and finally smooths the combined image to enhance the visibility of faint sources.

NOD_CHOP_SCAN

Reduction of nodded and chopped data taken in a scan pattern, specifically for Michelle. It is similar to NOD_CHOP, but will create a mosaic from nodded and chopped groups taken in a scan pattern.

REDUCE_FLAT

Reduces a flat field by masking bad pixels, subtracting a dark, and normalising the result. This recipe is for all instruments.

I.2 Modified recipes

• NOD_CHOP_APHOT   The photometry aperture has been increased to 5 arcseconds from 3, and inner and outer diameters of the sky annulus have been modified correspondingly. The reported photometry uses filter zero points and widths, the latter to find and report the broadband-$N$ equivalent magnitude and flux for narrowband-$N$ filters. Extinctions are now included.

  The recipe now determines the baseshift calibration, which measures the displacement of the source images from the nominal locations.

• ARRAY_TESTS   For Michelle the estimated readout noise is no longer reported.

I.3 Global changes

The main changes from a user perspective were as follows.

• Undefined or nonsense headers used by the recipes are replaced internally to pipeline, but not in the physical files, This usually permits the pipeline to complete. Recipes report the names of modified headers, for manual checking and correction of the raw data.
• Further sanity checks are included for Michelle. The waveform header is validated, and incorrect combinations reported. The mean data value must lie between 25000 and 48000 counts to prevent a warning message from appearing.
• Recipes make an on-the-fly night log in $ORAC_DATA_OUT. Night logs contain the group number. The file name has an _im suffix for Michelle.
• The default UFTI gain reflects the gain’s changing values arising from the use of a different controller.
• NDF blank titles are removed.
• The UFTI bad-pixel mask has changed, now having 0.3% more bad pixels.
• Centroiding is protected against data comprising all bad values.
• The hybrid registration has a further improvement for a special case.
• The near-infrared photometry uses the latest standards list, which also includes standards for the $IZJHKLM$ wavebands.
• In photometry recipes the sky units are now reported as counts/s/pixel.
• Scripts have improved error detection and reporting after accessing text files.
• The default display suffices have been augmented. These are for Michelle _cab, _scab, _cpc, and _rpc; and _dk for IRCAM.
• The calibration called mask, should be used to specify other than the default bad-pixel mask, rather than the bpm calibration.

The main changes from a programmer perspective were as follows.

• There is better structuring of the preliminary steps to permit instrument-specific variants, such as correcting the world co-ordinate system headers.
• The waveplate angle header values are standardised to a single format.
• The preliminary operation includes the removal of axes after a rearrangement in
  ORAC::Convert.
• The registration was modified to allow for changes to Ccdpack.
• The _COMBINE_CHOPPED_FRAME_ primitive as used by the NOD_CHOP family of recipes has new options. These comprise centroiding with validation, the method by which the individual chopped and nodded images are combined, the ability to apply block smoothing for faint sources, and the removal of pickup and bias variations. The removes column then row patterns, by subtracting the median of each column or row from all the values in that column or row.