[daip] Request for experimental Task

[Frazer Owen]

    Dear Eric et al,

    I have been working with GMRT continuum data for the first time.
I find that there is an artifact in the center of the field which limits 
severely
the noise. It looks a little like a correlator offset but is a bit messy.

    To explore the effect, I have averaged the residuals from my imaging
(after UVSUB) over the full 10 hours of the run. I then looked at the
results with POSSM by baseline and polarization. There are 124 channels.

    The result shows that on some baselines there is a coherent signal which
appears to be approximately constant. In general it has some spectral shape
in amplitude. Sometimes,it is a fast sine wave sometimes it only has a 
little
structure. Usually the phase across the 124 channels changes coherently.
Sometimes I see a less serious problem of a narrowband birdy as well.

    I would like to try to subtract this averaged signal from each 
visibility
in the database to see if that will remove the effect or at least reduce it.
However, there is no way that I can find to do this in AIPS. Thus I
am requesting an experimental task which allows me to subtract the
grand average of each baseline and channel from the time original
uv database used to make that average.

    One can think of many wrinkles on such a task for general use but
right now I would just like some way to try out making such a correction.
If this proves useful, then it would be possible to do something fancier.
Leonia is looking into it but maybe Eric can see some way to do this with
little or no work.