[fitsbits] Re: FITS for DSN spectral line observations

[William Pence]

Tom Kuiper wrote:
> 
> ... Please look at the first
> cut presented in http://DSNra.jpl.nasa.gov/devel/fits/
> ...

This is a quote from that web page:

    "Note that the order of the keywords TSYS-DIF and CALSCALE
    is significant. Performed in the order shown here, the 
    average baseline value will be zero. If the order is 
    reversed, the average baseline value will be the Tsys 
    difference between the "sig" and "ref" scans."

It is not a good idea to rely on the order of keywords in the FITS header. 
For one thing, it violates the FITS standard, which in section 5.1.1 states
"Except where specifically stated otherwise in this standard, keywords may
appear in any order".  From a practical point of view, there is no guarantee
that the order of keywords will be preserved if the header is copied to a
new file, and it is awkward with many existing FITS file interfaces (e.g.,
CFITSIO) for application programs to determine the order of the keywords. 

An alternative solution that would avoid these problems would be to define a
another keyword that would specify whether TSYS-DIF or CALSCALE should be
applied first.

Bill Pence
-
____________________________________________________________________
Dr. William Pence                          pence at tetra.gsfc.nasa.gov
NASA/GSFC Code 662         HEASARC         +1-301-286-4599 (voice)     
Greenbelt MD 20771                         +1-301-286-1684 (fax)