[daip] FITS header

[Mark Calabretta]

On Tue 2004/09/28 13:59:12 CST, Eric Greisen wrote
in a message to: Ian Pattison <ip at ast.leeds.ac.uk>
and copied to: daip at aoc.nrao.edu

>The for pixel 2975,3274 (a small source) aips gets
>
>01:32:56.70055,  30:42:17.7816
>
>Running ds9 I get at this pixel
>
>01:32:56.814,    30:42:22.94
>
>I tried changing the cdelts to the average (.332671) and then we get
>agreement.  Has aips got a probelm in interpreting CDi_j?

Eric,

The famous Eq. 191 of Paper II should be consulted.  It gives

  rho_a = 89.2985486
  rho_b = 89.2730479

Therefore this CDi_j matrix is not factorizable as a rotation followed
by a scale (or visa versa for that matter), so AIPS is in trouble.

As for ds9, the best I can do is compute the coordinates of your test
pixel (2975,3274).  Using WCSLIB I get

  01:32:56.8142, +30:42:22.9371

which agrees pretty well with the value you got from ds9.

Cheers, Mark