[daip] AIPS not reading FITS headers

[Eric Greisen]

landon at inaoep.mx writes:

 > I am having trouble getting AIPS to properly import a FITS file. The header
 > includes these listings:
 > 
 > RESTFREQ=  1.1527120400000e+11                                                 
 > VELO-LSR=  4.2000000000000e+05                                                 
 > DELTAV  = -2.6044000000000e+03                                                 
 > CTYPE1  = 'FREQ    '                                                           
 > CRPIX1  =  4.1826555060700e+02                                                 
 > CRVAL1  =  1.1527120400000e+11                                                 
 > CDELT1  =  1.0000000000000e+06                                                 
 > 
 > When I load in the file, AIPS imports the first axis's information, but does not
 > import the alternate velocity definition for the axis. I used ALTDEF in AIPS to
 > try to set the alternate value with the following inputs:
 > 
 > AIPS 1: AXTYPE     'RAD LSR '              'OPT' or 'RAD' plus 'LSR',
 > AIPS 1:                                    'HEL', or 'OBS'
 > AIPS 1: AXVAL      420000         0        Velocity at (new) ref. pixel=
 > AIPS 1:                                    AXVAL(1) + AXVAL(2).
 > AIPS 1: AXREF         257                 Reference pixel location
 > AIPS 1: RESTFREQ   1.152E+11   7.120E+07   Rest frequency in HZ =
 > AIPS 1:                                    RESTFREQ(1) + RESTFREQ(2)
 > 
 > This set the Delta V value to somethimg different than the value in the FITS
 > header. 
 > 
 > It would be nice to have AIPS read in the entire FITS header properly, but
 > possibly the format of the header is not fully compatible with AIPS. This FITS
 > file is of dimensions (FREQ,RA,DEC)=(512,3,3) and is compiled from 9 single dish
 > observations from Onsula Radio Observatory.
 > 

AIPS does read the header properly by previous standards.

The keywords you show above appear neither in any previous standard
nor in the current (proposed) standard.

Furthermore, you have noticed that the keywords above are also
incorrect.  ALTDEF computes the correct increment.

Eric Greisen