[daip] aips imagr

[Eric Greisen]

Cao Hongmin wrote:
> Hi Eric
> 
> As the VLBI dataset as concerned, what is the maximum image size (or how 
> many image pixels, eg. 1mas /pixel ), which aips can handle  and need 
> not to be corrected the primary beam attenuation and no-coplanar effect?

In general for VLBI, the field of view is tiny with respect to the 
single-dish beam areas.  So long as the phase stopping position is at 
the antenna pointing position, VLBI users have never worried about these 
issues (unlike VLA users).  The modern DiFX correlators for VLBI can 
correlate at multiple phase-stopping positions, so it is necessary to 
think about primary beam patterns for such data.  A new AIPS task, to 
adjust the CL table, is under testing at present for such data - note 
that the fields are so small in VLBI that a single correction for any 
particular phase-stopping position is all that is needed (perhaps as a 
function of time if the beam rotates on the sky and certainly a function 
of frequency).  To have a large field of view in VLBI would require 
integration times in the milli second range and bandwidths measure in 
Hz.  With normal bandwidths and integration times, time and bandwidth 
smearing will destroy the signal before primary beam and non-coplanar 
effects come into play.

These are basic interferometry matters and are better covered by, e.g. 
books from the Synthesis Imaging in Radio Astronomy books (ASP 
Conference Series, vol 180)

Eric Greisen