[daip] task FILLM

[Eric Greisen]

fgovoni at ca.astro.it wrote:
> Dear analyst,
> 
> I'm writing regarding the problem with the AIPS task
> FILLM for polarimetry data.
> 
> If I well understand the problem with FILLM regards only data
> performed after June, 27, 2007 and in such a case I have to use
> a correct version of AIPS (31DEC08 for example) to calibrate the data.
> 
> But, in the email you mentioned to a less serious problem also:
> "FILLM has also been calculating the weights for the same cross-hand data
> incorrectly, although this has a much less serious effect on the data."
> Could you better quantify such an effect please?
> To avoid this effect, do you suggest to use a correct version
> of AIPS (31DEC08 for example), both for new (after June, 27, 2007)
> and for old data?
> 
> Best regards
> Federica Govoni

It is hard to quantify the effect on weights.  We compute Q and U from 
RL and LR without use of weights, so all that might be somewhat wrong is 
the weight used in imaging Q and U.  With some form of robust uniform 
weighting, this should not matter much at all.  There might be a small 
effect in weighting data for determining the answers in PCAL.  But again 
that effect on old data is small.

The serious problem is with data observed after June 27, 2007.  For such 
data, you need to obtain a correct version of AIPS (31DEC07 with current 
patches or 31DEC08 current to now) and rerun FILLM.  You can copy the 
SN, CL, FG, tables from your currently processed files to the new ones.
Then run PCAL over again.

Eric Greisen