[daip] FARAD question

[Hanno Spreeuw]

Thanks,

I was inspired by Robert Braun's scheme for WSRT pol cal.
http://www.astron.nl/wsrt/WSRTsoft/cookbook.html

I thought I would use either this scheme or the five step approach to 
polarization calibration in the COOKBOOK, but replacing LPCAL/PCAL by 
SPCAL, because this is P-band and the leakages are frequency dependent.

I can avoid using FARAD before running PCAL/LPCAL/SPCAL because I have 
at least one unpolarized calibrator. Also, I can calculate the XY phase 
offset without correction for ionospheric faraday rotation because I 
have a linearly polarized calibrator with V=0.

So I just need to adjust the visibilities of the target in some smart 
manner before imaging Stokes Q and U.

Hanno.

Eric Greisen wrote:
> Hanno Spreeuw wrote:
> 
>> Hello, I have  a question about FARAD for WSRT data.
>> After changing the Stokes value from -5 to -1 in the header using 
>> PUTHEAD,
>> RL=Q+iU is now XY=-U+iV.
>> Assume V=0 for calibrators and target. XY=YX=real if the XY phase 
>> offset has been corrected for properly. Ionospheric Faraday rotation 
>> will then not affect the XY phase (~0), just the amplitude, if I am 
>> not mistaken.
>>
>> Also XX and YY are affected because Q is affected by ionospheric 
>> Faraday rotation. For the VLA, RR and LL are not affected by 
>> ionospheric faraday rotation if I understand correctly.
>>
>> Is it correct to infer that FARAD will not work for WSRT data because 
>> the XY phase does not change by ionospheric faraday rotation?
>> Is there a workaround? I mean, the RMs calculated by FARAD could still 
>> be used but would have to applied differently.
> 
> 
> I think it is safe to say that AIPS only works minimally with linear 
> polarization telescopes.  It can do some useful stuff with polarized 
> calibrators on WSRT data since they are equatorial.  But polarization 
> calibration itself would be beyond our capabilities and of course FARAD 
> cannot do something correct for linear polarizations when the header has 
> been buggered to lie about the polarization...
> 
> If you know someone who is expert in linear polarization who would like 
> to work in AIPS, we might be able to find some funding....
> 
> Eric Greisen