relative fluxes

[Petr Skoda]

I have to comment on this as well :

> format (1D image) is passed on as "NATIVE" within the VO, and because
> of that the VO (not the final users which are happy with it) does not know
> how to treat it. It would be much simpler to recognise that such ground-based 
> format
> exists and that it is widely used, and hence allow proper VO description of 
> it.




>
> With a little addition to the IVOA standard we would suddenly open the doors
> to many ground-based observatories!

It is exactly the deadlock here - there is a wealth of spectra not 
"published" by the server but available for download in this FITS format 
from various ftps and small observatory web pages. People want to let the 
world know their work but they do not know how to do it properly ....
See further on...

>> DM and its formats now that we have this.  We can down-convert on a
>> case by case basis to support legacy apps if necessary, but with the
>> introduction of the VO why not try to move to another level.

The case by case basis would cover 90% of potencial resources for ground 
based SSA

> is much easier than having to transform all the existing ground-based 
> spectra.
> Plus, some of those tools already support the 1D image format anyway.

To be precise I think that tho only spec tool that does not understand it 
is VOSpec - SPLAT and SpecView can understand well the basic 1D FITS.


> It would indeed be nice to know which other data providers use the 1D FITS 
> IMAGE format
> for their 1D spectra in their archives...

In fact most of the spectra from smaller telescopes are reduced in IRAF 
(echelle package or onedspec/twodspec ) or 
MIDAS.  Even those telescopes  having special pipelines (ELODIE, Espadon, 
HARPS, UVES ) will produce at the end the 1D FITS output as it is 
supported by many legacy applications and can be analyzed directly.
in case of IRAF there is a possibility to staore original nonlinear 
dispersion in headers in WAT keywords, but I know only very few people 
using it. Most reductions are finished by rebinning to equidistant
wavelenght using the CRVAL and CDELT1 FITS keywords.

I have seen examples of such a spectra being produced e,g, in DAO,
output from Kitt Peak, CTIO, Apache Point telescope, Bulgaria Rozhen 2m,
Tautenburg 2m, GAO and Okayama archives in japan, TNG, WHT, INT {La Palma)
Crimean spectrographs, BOES in South Korea, SAAO 2m in Sutherland, (low 
dispersion spec and the echelle is converted to 1D on request), probably 
SALT spectra, Hercules spec at MJO New Zealeand, all Australian spectra - 
MSSO, AAT echelle and MOS, spectra from Poznan echelle, HEROS spectra from 
Heidelberg (including ours when HEROS was in Ondrejov), ELODIE, SOPHIE 
(2m at OHP), Calar Alto spectorgraphs, Pic Du Midi spc, Asiago observatory 
echelle, La Silla Danish 1.5m and 2.2. spectra from FEROS echelle, LBT, 
MAGELLAN, MMT, MT Palomar 5m, SPM mexico  spectrographs,

Those are examples of most ground-based observatories with large spectra 
archives. Unfortunately they are not data providers (yet). Some of them 
are not having capacity to create archive, some are not willing to do so 
(e.g. Keck except the HIRES spectra). But I think most of such 
observatories would like expose (part of) spectra to the world (as it may 
risen the reputation and acknowledge the institution) but they do not want 
to do  complicated development. In fact they have to reduce the spectra 
and calibrate it and the astronomers want to analyze them (and majority is 
using IRAF splot or MIDAS Xalice in adition to many custom tools (that 
undertand only ASCII table and 1D FITS ).

So if you convice them to publish the data it must be done easily - get 
some bunch of FITS files, add some metadata using FITS header and put 
somewhere on web.

The part of the above list is my personal experience (I worked with the 
data) or by exchange of results (already reduced and calibrated) by my 
colleagues and friends. Some spectra were given as examples on web page of 
the observatory and at some I have seen just header and at soem I have 
asked about the reduction pipeline and the anser was standard IRAF 
packages.

In addition to this the amount of so called spectral libraries is giving 
just FTP to download 1d FITS spectra .....

(example is summarized e.g. here 
http://www.ucm.es/info/Astrof/invest/actividad/spectra.htm)
Most of this wealth of spectra is waiting to be put in VO (e.g. some of 
them may go through CDS as electronic attachments or just by FTP 
reference)


The bottom line question is : what kind of collaboration is between 
organizations with strong VO development and such a smaller telescopes at 
the national level ?   (e.g. DAO spectra and CADC, VAO and Keck, Kitt 
Peak, CTIO telescopes,
....) but that's another question to be solved elswhere.


I hope it is now more clear what I mean by ground based archives and what 
amount of data it represents.



*************************************************************************
*  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                                                       *
*************************************************************************