[daip] keyword for aberration correction

[Walter Brisken]

The main thing is to document the contents of the file, but I think that 
we need to think if UVFIX should respect this parameter -- I think it 
probably should not attempt to recalulate UVW from first principles for 
data with ABERCORR set, perhaps with some override capability.  The 
corrections UVFIX makes would not be correct for near-field observations I 
believe.  A similar issue will arive with EVLA as I expect they will be 
correcting UVW before writing archive records, but their corrections will 
probably be APPROXIMATE.

-Walter

On Thu, 17 Jul 2008, Eric Greisen wrote:

> Walter Brisken wrote:
>>  Today at the VLBA Test and Development meeting I mentioned that with the
>>  software correlator I plan to use UVW coordinates that are corrected for
>>  the various relativistic effects (aberration, etc).  I believe this should
>>  be made clear in a table keyword.  I would like the keyword (call it
>>  ABERCORR here, subject to other suggestions) to have at least three
>>  levels:
>>
>>  ABERCORR = UNCORRECTED -- the uvw values are those returned by calc
>>  ABERCORR = APPROXIMATE -- a correction, acturate to some level, is made
>>  ABERCORR = EXACT -- all corrections consistent with the delay model are
>>  made
>>
>>  Does this sound reasonable?  I would suggest making this an optional
>>  keyword in the UV_DATA table of FITS-IDI files.  The default if left
>>  unspecified would be UNCORRECTED.
>
> Did you have any intended use for this keyword other than informative?
> UVFIX does its thing and does not care how good the old u,v,w are and the 
> rest of AIPS takes them at face value in any case (having no other choice).
>
> Eric
>
>