[daip] New Staff Reply - [!NTC-279643]: EVLA HI Spectral Line Calibration in AIPS

[Eric Greisen]

New Staff Reply: EVLA HI Spectral Line Calibration in AIPS

1. The UVFITS format comes through CASA even if all done by
the archive.  CASA does seem to write simple data sets 
correctly for the EVLA although without some of the tables
like the SY one.  (ALMA data are messed up in this process
for some reason and CASA has been slow to address the issue.)
In general calibration with the SY table may remove the need
for a gain (3C286 et al.) calibrator at all.  But, at L band,
the RFI environment is so bad that the SY table is compromised.

2. If you have a traditional flux calibration source, then REWAY
is a good way to set the weights and the SY table would need a
lot of work at L band to be useful.  Thus calibration done your
way probably works just fine.  At higher, more benign, frequencies
the SY table route appears to be quicker and better but even there
not totally essential.

3. RFLAG is new and people are still learning how to use it.  The
task does seem to work well and to save an enormous amount of time
over SPFLG and TVFLG.  The consequences of such a blind flagging
however are still not understood and there may be some negative
ones.

Reasonable people often disagree over the best way to process data
and there have even been rather heated arguments around here.  I
have written what I think is a good general plan which mixes both
some current reality of the EVLA and some hoped for future reality
in a plan that should be not a rule but a suggestion.  Different 
kinds of data have different problems and require different approaches.
If your data are going well with your scheme then don't worry about
Appendix E suggesting something else.  (But do experiment a bit with
RFLAG - it may save you a lot of work.)

Eric Greisen


Ticket Details
===================
Ticket ID: NTC-279643
Department: AIPS Data Processing
Priority: Default
Status: Closed
Link:  https://help.nrao.edu/staff/index.php?_m=tickets&_a=viewticket&ticketid=1831