-
Type:
Task
-
Status: Done (View Workflow)
-
Priority:
Major
-
Resolution: Done
-
Labels:None
The 2d pipeline needs a seed solution for fiber positions and wavelengths, both to identify the fiber traces and to initialize the wavelength solution for arc line matching. The simulator knows the geometry from the input jemax spot files and is currently supplying stella with a one-off (fiber#, y) -> (x, lambda) FITS table.
We would like to use proper datamodel files to encapsulate the table. There is an existing pfsFiberTrace file which can hold the (fiber#, y) -> x mapping, but it does not have space for wavelength info.
Surely we want to persist wavelength solutions outside of pfsArm files, since the maps will get applied to several image files.
Not sure whether we should:
1. create some pfsFiberWaves file, which just contains the (fiber#, y) -> lambda map.
2. beef up the existing file by adding an optional HDU containing a wavelength column for the PHDU table.
Now that I write it down, 2 looks too gross. So 1? Shall I?