NOAO Science Archive Domain Model contribution

[Phillip Warner]

Greetings to all in Data Model land,

I have been on this list for a while now, keeping track of the
progress that is being made by the Data Model group.  I appreciate all
the hard work that has gone into defining the current form of the data
model, as I found that the models are closely related to the model we
have developed for the NOAO Science Archive (NSA).  I would therefore
like to contribute a snapshot of our current domain model.

NOAO has been developing a system that will eventually replace and
extend the current NSA.  For those that are not familiar with the NSA
(http://archive.noao.edu/nsa), it is an archive of reduced and highly
processed ground-based optical and IR data produced by NOAO Survey
teams (see http://www.noao.edu/gateway/surveys/programs.html for past
and present surveys).  The NSA, in its current state, holds only
survey data.  The new development effort, however, is focused on
creating an archive for raw, calibrated, and reduced data, as well as
catalogs of astronomical objects and extracted spectral and time
series data, in the O/IR domain.

The NSA is an ongoing development project. We are currently working on
developing the needed infrastructure to archive raw data streams
coming from NOAO telescopes.  Hence, this domain model is not yet
complete; some areas are more defined than others.

The current version attempts to capture elements of an astronomical
observation, much like the Observation model.  Also, we have
incorporated, albeit rather superficially, the concept of Quantity
into our model.  We have also tried to make our model as consistent as
possible with the current Observation data model, although ultimately
there is not a direct one-to-one mapping for many elements.  An
example of this is our use of the term "Observation", which in our
project's glossary, is regarded as a synonym for "astronomical
exposure".

Although much of this domain model is directed to the internal needs
of the NSA as a data center, we think that many parts apply generally
to what is needed for the VO model.  Furthermore, we think that this
model could provide some insight into how the VO could work with each
individual institution.  Of course, we do plan for interoperability
with the VO, and therefore we find that providing our modeling to the
DM group may be helpful.

I have posted this document on the IVOA wiki, in the DM group web:

   http://ivoa.net/twiki/bin/view/IVOA/IvoaDataModel

or more specifically,

   http://ivoa.net/internal/IVOA/IvoaDataModel/NSA-DM-v0.4.0.pdf

Thanks for reading and please enjoy,

Phillip Warner
NSA Team
NOAO