[daip] velocity header definitions

[Lynn D. Matthews]

I've noticed a peculiar phenomenon in regards to the velocity
definitions given in the headers of some of my VLA data.

I currently have on disk 5 datasets taken on 5 consecutive days in
essentially identical observing modes, where a "barycentric"  velocity
reference frame was requested in the OBSERVE file. For 3 of the datasets,
the header claims that the "vel type" is "OPTICAL wrt YOU" (i.e., VELREF 3
0) and the velocity axis of the cube (after running ALTSW) is labelled
"FELO-OBS". For the other days, "vel type" is "OPTICAL wrt SUN" (VELREF 2
0) and the velocity axis is labelled "FELO-HEL".

At first I thought that somehow my reference system for these former data
had gotten changed to an Earth-centered one, but calculations show that my
center frequencies are just as expected for a barycentric reference frame,
hence this header information is bogus.

It turns out that there is actually one minute difference in the OBSERVE
files for those two days; on the two days with the faulty header
information, I had the "qualifier code" set to 0, whereas in the other
days it was set to 1. Could this qualifier code be triggering an
erroneous redefintion of the velocity frame within FILLM?

This also brings up a more minor point, which is that a heliocentric rest
frame is actually not an option for VLA observations. In pratice, the
difference between barycentric and heliocentric is negligible given our
current velocity resolution, but labelling the two interchangably does
have the potential for some confusion.

Lynn