[Obscore DM update ] Redshift axis and Spectral Axis

[Louys Mireille]

Le 12/05/2015 18:29, Patrick Dowler a écrit :
>
> Is the addition of a doppler/redshift axis to ObsCore something we are 
> talking about for ObsCore-1.1 or later?
>
Dear all ,

This is a good question .
I think there is some misunderstanding for the scope of the various data 
model versions.

After many discussions in the last few months , trying to figure out 
practical examples of data sets  described in Obscore
and accessed via ObsTAP, we realized there are various use-cases where 
the WCS information is tricky to expose in the simple tabular format of
an ivoa.Obscore table.

The suggestion I had in Banff to add a redshift axis is not satisfactory 
for
the simple discovery step that ObsTAP supports , as it adds 9 or 10 
keywords
that will be null for more than 80% of the datasets served.
Numbers to be checked, but at least a large fraction of observations are 
not velocity data.

The problem with redshift axis is to find the appropriate level of 
description
and consider which information is enough just to discover a velocity 
cube , for instance.

For the next update of ObsCore , that is 1.1 , we will have only the 
possibility to discover if a dataset has velocity content,
by using the "em_ucd" field.

For the future updates of Obscore, a full description, with related 
reference frequency, position , etc , can be defined and prototyped.
However the Image DM already contains the full description of the 
Redshift axis, so specific data collections with velocity measurements
can be delivered in an SIAV2 getmetadata() capability, based on ImageDM.

Finally after much hesitation and steps forwards and backwards, I 
suggest to have a minimal description of the redshift content in one keyword
for Obscore 1.1.
Then for a general search, the selection criteria **for discovery** stay 
compatible to the original ObsCore strategy :
em bounds in meter , ICRS for positions , etc ..

For data retrieval, the d_ features exposed  previously can be used in 
extra tables, or the ImageDM serialisation.

This keeps ObsTAP focused on discovery , and allows richer description 
from ImageDM to be used via SIAv2.

I hope this helps,
Mireille











-- 
Mireille Louys	, Maître de conférences
Centre de Données ( CDS)		Icube & Télécom Physique Strasbourg, Pôle API
Observatoire de Strasbourg 		300, boulevard Sébastien Brant
11, Rue de l'Université			CS 10413
67000 Strasbourg 				F - 67412 ILLKIRCH Cedex
http://astro.unistra.fr			http://www.telecom-physique.fr
tel : 03 68 85 24 34

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ivoa.net/pipermail/dm/attachments/20150512/0f7d2744/attachment-0001.html>