[daip] New Ticket - [!SJM-103823]: extra Stokes I columns in uv data

[Lynn D Matthews]

New Ticket: extra Stokes I columns in uv data

I have imported into 31DEC09/31DEC10 AIPS some model MERLIN data for comparison with real observations, and have been confounded by a problem that I've not previously encountered.

The files read in without any errors, and if I plot, e.g., amp vs. time using VPLOT or print the visibilities using UVPRT, everything looks completely sensible. However, if I image these data, I find strong residual PSF-like structure that will not clean. Upon examining the visibilities with PRTUV (instead of UVPRT), I discovered that there are three extra "hidden" Stokes I columns, each with different values of amplitude and phase.

I am guessing that IMAGR is reading these extra columns, but where are they coming from?

I have verified that the models are generating only a single polarization, and so suspect that the problem is related to the FITS formatting. The model data are written out in the old random group format; should this still be backward compatible? Or could the problem be some missing keyword or information?

Thanks.

Ticket Details
===================
Ticket ID: SJM-103823
Department: AIPS Data Processing
Priority: Default
Status: Open
Link:  https://help.nrao.edu/staff/index.php?_m=tickets&_a=viewticket&ticketid=186