SED Data Model: Questions and Comments

[Jonathan McDowell]

I'm still working on V0.93 of the model, hope to have it
up soon.

On Feb 3, Igor commented on my ruminations on errors.

: On Wed, 2 Feb 2005, Jonathan McDowell wrote:
: > 2. No flux calibration.
: > I'm still thinking about this one. In your case (shape right,
: > flux cal absent) one could just put a huge systematic error
: > bar on, but it would be better to have a special tag.
: 
: I don't think a huge systematic error bar is a good solution,
: because somethimes even the order of magnitude is unknown.

Yeah, good point.


Gilles Duvert shouted:

> A measure is a value and an error. ...
> DO NOT PERMIT .. WRONG DATA TO BE OUT IN THE VO

But that's not the case. Data with a good wavelength calibration and
no flux calibration still allow reasonable measurement of the wavelengths
of spectral features and are therefore scientifically useful - many papers
have been published on such data. The key is to find the right way to
describe them, rather than to forbid them.


: The question connected to the flux calibration is handling "observed" fluxes
: (FLUX-OBS in FITS) -- that are obtained with the ground-based observations,
: and "physical" (FLUX-PHY) fluxes -- ones outside atmosphere. The correction

I missed this - where are FLUX-OBS and FLUX-PHY defined? In the WCS papers
somewhere?

: > 3. Atmosphere and vacuum wavelengths
: > what extra metadata do you need to let applications make the correction
: > from air to vacuum? Airmass, temperature, humidity, ...??
: 
: There is an IAU standard (!), that doesn't care of meteo params.
: It tells (Morton 1991,ApJS,77,119) that
:       AIR = VAC/(1.0 + 2.735182E-4 + 131.4182/VAC^2 +  2.76249E8/VAC^4)
: where AIR and VAC are lambda-s in Angstroms. Should we follow it, or we should
: implement something better (if exists)?

Sounds like a good start, but when converting 
we should make sure to add metadata like
  <air_vac_algo>Morton1991</air_vac_algo>
so that we can trace what algorithm has been applied.

 - Jonathan