[daip] Polarization Calibration

[Eric Greisen]

Doug Roberts wrote:
>   Hi Eric,
> 
> I have a pretty complex pipeline calibration script that I'd like to run 
> past you to make sure I'm doing polarization calibration correctly.  The 
> script does Stokes I as well as polarization calibration for our 
> Galactic center projects.  I'm using RLDLY, PCAL and RLDIF.  One 
> question before I send the the big script for review is about the 
> ordering.  In the documentation for spectral polarization calibration to 
> to the following steps:
> 
>    1. RLDLY
>    2. PCAL on 3C286 - CONT
>    3. RLDIF using calibrator observed throughout run for parallactic
>       angle coverage - CONT
>    4. PCAL - SPECTRAL
>    5. RLDIF - SPECTRAL
> 
> My script does the following
> 
>    1. RLDLY
>    2. PCAL on 3C286 - CONT
>    3. PCAL - SPECTRAL
>    4. RLDIF using calibrator observed throughout run for parallactic
>       angle coverage - CONT
>    5. RLDIF - SPECTRAL
> 
> Are these equivalant or should I do all the continuum before doing the 
> line polarization?
> 
> Thanks in advance!
> Doug
> 
> 

I am a little perplexed by these.  Normally PCAL is run on the source 
with the wide parallactic angle coverage (and unknown polarization) 
while RLDIF is run on 3C286 because its polarization angle is 
essentially constant and well known.  RLDIF is also usually run only on 
a single scan, if it is long enough to have decent S/N.  If one is going 
to do the SPECTRAL solution, then the continuum solution has little 
value other than confirming that the spectral solution is not crazy.


Eric Greisen