jwst
jwst copied to clipboard
spacetelescope
Add outlier rejection parameter file for MIRI MRS
Issue JP-1470 was created on JIRA by David Law:
During recent discussion about creating parameter reference files for the Outlier Rejection step we noted that 'parameter reference files' containing default values for parameters and selected via crds as a function of filter/band/etc will not work for the MRS. Since MRS contains multiple bands in a single file, even when outlier rejection is run on a purely band-dependent basis it would require multiple such parameter reference files (e.g., when processing a 12A exposure it would need information for both 1A and 2A).
A cleaner solution is to have a single 'reference file' for MRS outlier rejection in which the appropriate values to use for all bands are contained in a single place and can be used as necessary. This follows the approach already in use for cube building using the values in the 'cubepar' reference file which are too complex for a crds-selected 'parameter reference file' structure.
This will require creation of a new reference file data model, and code changes to outlier rejection to read the relevant values from this file. Such changes would allow removal of the hard-coded value kludge currently in place for the MRS.
A draft reference file is attached.
I don't think this ticket is relevant any longer; based on flight data we're currently using an ordinary parameter reference file for outlier detection. There's work still to be done to quantify whether these parameters can be improved, but I think it's best if that were a new ticket instead of keeping this pre-flight ticket open indefinitely.