[daip] FITS header

[Eric Greisen]

Ian Pattison writes:
 > Eric,
 > 
 > One thing I have noticed is that in the AIPS header, the coordinate
 > increment for RA and Dec are -0.326702 and 0.338639 asec respectively. The
 > FITS header states 0.333 asec/pixel. Any idea why AIPS is saying this? I 
 > know I could edit the values using the keyword 'CRDELT' in the task 
 > PUTHEAD.
 > 

  AIPS has solved the CD1_1, CD1_2, CD2_1, and CD2_2 matrix for the
two increments and the rotation.  It is a correct solution - if you
look at those values you will see that the two increments cannot be
equal.

ERic Greisen

 > Regards
 > 
 > Ian
 > 
 > 
 > On Tue, 21 Sep 2004, Eric Greisen wrote:
 > 
 > > I have looked at the header and AIPS seems to be doing the right thing
 > > - the only standard WCS cards are all at the end and they say what is
 > > showing up in the header.  There are other non-standard wcs-like cards
 > > that indicate an RA near what the standard one shows but a Dec that is
 > > 4 arc min different.  That is about 700 pixels - not anything special
 > > such as pixel 1,1.  I see no coord info in the header at all that
 > > would be with 10 arc sec of the position shown.
 > > 
 > > This is basically a transposed image with Dec varying fastest along
 > > the first axis (rotation ~90 deg).  In aips this would be a rotation
 > > of ~0 and the Dec axis shown first.  There may be a bit of skew that
 > > aips cannot handle.  The standard keywords look tacked onto the end
 > > without comment - makes me wonder how they got there and how reliable
 > > thay might be.  
 > > 
 > > Eric Greisen
 > > 
 > 
 > -- 
 > ******************************************
 >  Ian Pattison 
 > 
 >  Astrophysics Group 
 >  School of Physics & Astronomy
 >  University of Leeds
 >  LS2 9JT
 >  United Kingdom
 > 
 >  Tel: +44 (0)113 3433872
 >  email: ip at ast.leeds.ac.uk
 >  WWW: http://ast.leeds.ac.uk/~ip
 > 
 > ******************************************
 > 
 >