[daip] combining epoch 1950 and 2000 data

[Leonia Kogan]

Hi Tyler,

>How exactly do I
>get UVFIX to change from epoch 1950 coordinates to 2000?
>Do I use the adverb SHIFT with the appropriate distance
>in arcsec between the 1950 coordinates and what they 
>would be in 2000 coordinates?

No you do not need to use SHIFT. Put SHIFT=0. 
UVFIX calculates UVW from scratch. If the original UV data are given at J2000
coordinate system then it uses this coordinate system for the calculation.
If the original UV data are given at B1950 coordinate system then it 
recalculates the source position to J2000 coordinate system and uses J2000 
coordinate system for the calculation of UVW.

UVFIX doe not change the data itself.

The main effect of change of the coordinate system is rotation. So if you 
compare the images of the same object created using UV data given at B1950 and
J2000(or any other system, apparent coordinates for example) you should see 
the rotation of the image.

That is why UV should be given always at the same coordinate system if you want 
to compare observation made at diferent time.

That is why DBCON makes a right decision rejecting concatanation of the
two UV data at different coordinate system.

I have checked UVFIX here recalculating UV data given at B1950.
As i expected the output file has J2000 coordinates of the source and 
AIPS 1: Coordinate equinox 2000.00.

Some more comments:

1. The UV data have to be a single source file. Usually UVFIX is used 
   after calibration applied by SPLIT/SPLAT
2. UVFIX calculates UV taking into account some second order effects which 
   are important for precise astrometry

Leonia


---- Begin Included Message -----

>From nordgren at lowell.edu Fri May 25 11:18 MDT 2001
Date: Fri, 25 May 2001 10:18:14 -0700 (MST)
From: "T. Nordgren" <nordgren at lowell.edu>
To: lkogan at zia.aoc.NRAO.EDU
Subject: Re: [daip] combining epoch 1950 and 2000 data
X-Sun-Charset: US-ASCII
Content-Type: text
Content-Length: 553
X-Lines: 16
Status: RO


Dear Leonia,
	Thank you for your email. How exactly do I
get UVFIX to change from epoch 1950 coordinates to 2000?
Do I use the adverb SHIFT with the appropriate distance
in arcsec between the 1950 coordinates and what they 
would be in 2000 coordinates? It seems to me from reading
the explain-file that this will just move the galaxy away
from the phase center which isn't what I think I want to
do. Running UVFIX with SHIFT 0 0 doesn't do anything to the data
as far as I can tell (i.e the output header files still says
epoch 1950).

Cheers,

Tyler


----- End Included Message -----