[fitsbits] uncertainties in (WCS) FITS headers

[Saskia Prins]

Hello,

Is there a more or less standard way of recording uncertainties in FITS
headers.  Especially for WCS headers, I would like to be able to include
into my institute's observing data, an uncertainty estimate on the
positional information that is understood by data reduction packages, ie
something more widely recognized than a local MYERROR-like keyword.

Also I would like to be able include in the headers whether a nominal
first-guess was applied or a proper astrometric or wavelength solution
solution was obtained.  A type of quality-flag on the WCS information,
other than a local header.

A practical example: in imaging data fresh from the telescope, I would
like to add header information specifying this is a nominal WCS with
error estimates on the CRVALn information.   

Equally, in spectroscopic data I want to make clear whether we are
dealing with an estimate based on grism parameters and some assumption
on slit position or with a full wavelength solution based on arc-lines.

Does anyone know of conventions and keywords to explicitly record
these types of information?

Regards,
Saskia Prins