[daip] [forwarded from Riccardo Cesaroni] UVLIN

[Patrick P. Murphy]

------- start of forwarded message (RFC 934 encapsulation) -------
Content-Length: 1451
From: Riccardo Cesaroni <cesa at arcetri.astro.it>
To: pmurphy at NRAO.EDU
Subject: UVLIN
Date: Thu, 16 Mar 2000 17:04:10 +0100 (MET)
Message-Id: <200003161604.RAA06650 at hagen.arcetri.astro.it>

Dear Pat,
sorry to bother you from time to time, but I've found out something that
might be important for aips users reducing spectroscopical data.
Long ago (1992 or so) I reduced some NH3 line data: in order to remove the
continuum I used both UVLIN and IMLIN. At first I was convinced that the
result was identical and I decided to use the cube obtained with UVLIN.
However, now I realize that in one of the sources the map obtained with UVLIN
is shifted by about 0.6" in RA with respect to the one made with IMLIN.
I came to the conclusion that this is due to the fact that when running
UVLIN on this source, I used a non-zero SHIFT in RA: for the other sources,
instead, I set SHIFT to 0 and the maps seem OK. Note that the shift in the
map is 0.6", but I used SHIFT=0.96 0 in UVLIN, which with a source declination
of -20deg corresponds to a true shift of 0.90": in other words, the bug cannot
be explained in a trivial way by assuming that UVLIN doesn't undo the shift
(as it should, according to the manual).
Although I cannot be 100% sure that UVLIN was indeed responsible for the
error on the map position, I cannot see any other sensible explanation.
Note also that the continuum map obtained by averaging line free channels
is OK.
I'd like to know your opinion about this. Perhaps this is a well known bug
which has been fixed long ago. In this case, I apologize for asking silly
questions!

	      All the best

			      Riccardo Cesaroni
------- end -------