relative fluxes

[Petr Skoda]

>
> We basically agree then. For ABSOLUTE and RELATIVE (as per Spec DM
> definition) it is clear what to do in a VO application and the vertical
> scaling could be enough.

>From my point of view, yes.
But (just checking the SDM 4.6.5) It seems to me the definition of 
RELATIVE in fact nonsense in mathematical  treatment.
I cannot imagine the "systematic error" (except simple scaling) that ratio 
of two values is correct:
If it is additive error (zero point shift) than it is not true.


It is exactly problem of relative depth of spectral line on normalized 
(rectified) spectrum in case of background scattered light.

F(line)/F(cont) is not same as (F(line)+B)/(Fcont+B)

with increasing B the depth of line decreases (lowering contrast)

So the RELATIVE definition does not work in this case.


> In particular, the "RECTIFIED" (divided by a, most of the times unknown,
> function) is currently under the uncalibrated tag that also covers
> counts, normalized to the continuum, and many things more...

I had the feeling that NORMALIZED in sense of rectified as defined 
in SDM (divided by reference spectrum ucd=arith.ratio, units blank)
where the reference spectrum should be 
phot.flux.density;em.wl;spect.continuum  in erg cm-2 s-1 A-1 (arsec-2 
seems to be inapropriate here)

But this would in 99% be usefull only for cases when you can store the 
continuum description (it would be nice in clients to switch on and off 
the continuum normalization).

But for the practical reasons I would now reserve the word NORMALIZED just 
for continuum normalized spectra  as this is the prevailing kind of spectra 
representation in ADS and final output of most high resolution spectra 
processing.

If client will select only NORMALIZED spectra they can be easily 
overplotted and stacked to see profile changes.

So all other are UNCALIBRATED (and the proper UCD may depend on nature -
probably in counts (most of basic pipeline processed products).

Petr



*************************************************************************
*  Petr Skoda                         Phone : +420-323-649201, ext. 361 * 
*  Stellar Department                         +420-323-620361           *
*  Astronomical Institute AS CR       Fax   : +420-323-620250           *
*  251 65 Ondrejov                    e-mail: skoda at sunstel.asu.cas.cz  *
*  Czech Republic                                                       *
*************************************************************************