[fitsbits] CROTA keyword

[Mark Calabretta]

On Mon 2003/07/07 15:34:39 GMT, William Thompson wrote
in a message to: fitsbits at donar.cv.nrao.edu

>I've encountered some FITS files which contain the keyword CROTA to denote a
>rotated 2D image.  My reading of the FITS standard (forgetting WCS for the
>moment) is that there really should be two keywords, CROTA1 and CROTA2, one 
for
>each axis.  Is that correct?

CROTA without an axis number is incorrect, it violates the syntax for this
reserved keyword defined in WGH (A&AS, 1981, 44, 363).

However, no semantics were defined for CROTAn in WGH.  The only subsequent
use that we know of was that of the AIPS WCS convention wherein only one
CROTAn card was given.  It was associated with the latitude-like axis
(usually CROTA2).  See Sect. 6 of WCS paper II (A&A 2002, 395, 1077) for
a full discussion.

Mark Calabretta
ATNF