relative fluxes

[Randy Thompson]

Thanks for defining this term. I think my comments still
apply though.

 - The "1-D FITS" format you describe is one of numerous formats that
   have been used for storing spectral data. I can't argue about its 
popularity
   among ground-based observatories though.
  - More importantly, in my opinion it will take more work to generate the
   needed metadata than to reformat the FITS files. That work will be 
required
   whether the metadata is to be added as FITS keywords or just written 
to the
   VOTable response.

I guess its left to be seen whether officially supporting this format is 
enough
incentive for observatories to register new SSAP services. If it is, I 
would support
it, although I think the metadata should be required in any case.

By the way, I don't think the FITS format described in the SDM was 
selected because
it was a common FITS format. It was choosen because it was a logical 
format for storing,
and describing, spectral data, and in theory,  could be expanded to 
support SEDs, light
curves, echelle spectra, etc.
 
Randy

Alberto Micol wrote:
>
> Hi Randy,
>
> I meant the format like for:
>
> http://www.ucm.es/info/Astrof/fgkmsl/hd185395_hb_5.fit
>
> (thanks Petr! very good collection of examples!)
>
> where the primary header unit shows something like:
>
> NAXIS = 1
> NAXIS1 = 1024
> CRPIX1 = 1
> CRVAL1 = 4500 /[angstrom]
> CDELT1 = 2.5 /[angstrom]
>
> and the spectrum is stored in a 1d array of linear wavelengths.
>
>
> Though is clear that additional VO metadata would be needed, that could
> done at the level of the VOTABLE, leaving the FITS unchanged.
>
> Thanks,
> Alberto
>
>
>> On 29 Jun 2009, at 16:17, Randy Thompson wrote:
>>
>> Alberto Micol wrote:
>> I can report that ESO stores spectra in 1D image format.
>>
>> It would indeed be nice to know which other data providers use the 1D 
>> FITS IMAGE format
>> for their 1D spectra in their archives...
>>
>> Hi Alberto,
>>   I've seen the term 1-D FITS used a lot lately, but have never seen 
>> a definition. Does it refer to
>> a file with an array of fluxes stored as a primary array and (linear) 
>> wavelengths defined with keywords
>> (e.g., WUPPE)? Does it include data where an array of wavelengths are 
>> stored in one file and an
>> array of fluxes in another (e.g., FOS, GHRS)?  Does it includes files 
>> where arrays of flux, wavelength,
>> error estimates, and sigmas are all merged together into one 
>> non-homogeneous primary array  (e.g., HUT)?
>> I think you may be grouping a family of file formats under the title 
>> "1-D FITS".
>>   We have archived files within MAST in all these formats and more, 
>> and is the reason we are
>> creating VO-compatible versions. The conversion of the data itself 
>> was not difficult using existing IDL
>> routines. The difficulty is usually in calculating the metadata 
>> necessary to allow data from one mission
>> to be compared with that of another;  something that needs to be 
>> provided with any file format.
>>
>> Randy
>