ASOV 17 Jan 2012

Development of a Solid Spectroscopy Data Model (SSDM)

“Grenoble Astrophysics and Planetology Solid Spectroscopy and Thermodynamics" database service

Bernard Schmitt, Damien Albertand the SSDM Expert group*

Institut de Planétologie et d’Astrophysique de Grenoble, CNRS / UJF

SSDM General Structure

SAMPLE INSTRUMENT

EXPERIMENT

SPECTRA

REFERENCES

BAND LIST

SSDM general structure

PLANNING 2010 « Spectrodatabase »

• Relational database advantages– Permits to store a wide range of parameters

variations in experiments.

– Experiment samples can handle complex structures (multiple layers / materials / constituents / precursors).

– Samples and experiments can have a relation to a parent sample/experiment, enabling full history for those.

– At term for an advanced use every parameters could be searchable.

• Molecule database identification standards available in SSDM :

– InChIKey (International Chemical Identifier).

– CAS Registry Number (Chemical Abstracts Service).

– IUPAC Name

Solid Spectroscopy Data Model

SSDM Overview

PLANNING 2010 « Spectrodatabase »

• Spectrum « meta-data » in single table• Spectrum data stored in database• Advanced Spectrum products• Additional spectrum informations

displayable

Spectrum data model

Spectrum tables

Absorption coefficient tables

Sample description

LAYERS

LAYER 2

LAYER 1

MATERIALS (grains: simple or complexe)

mixing

CONSTITUENTS (phases)

arrangement

SPECIES (molecular, mineral, atomic)

combinations

- pure solid, - molecular mixture, - polymer, - clathrate, hydrate, …- adsorption, absorption, …

- heterogeneous polycrystal, - coated grains- crystal with adsorbed molecule

- complex mix of constituents, …

SAMPLE

Sample scientific description

PLANNING 2010 « Spectrodatabase »

• Heavy relational tables• Handle all of the scientific description

specificities

Sample & Material model

PLANNING 2010 « Spectrodatabase »

• Parameters set are a way to store the same meta-datas for different objects

• Static values are stored in the object table

• More versatile solution, but difficult to handle simply with datamining tools

• Necessity of an interoperability layer with communication protocols

Parameters set model

PLANNING 2010 « Spectrodatabase »

• Ensures valid XML documents• Human-readable errors

Import XML Schema

PLANNING 2010 « Spectrodatabase »

• Can be produced by hand or automated

• Doesn’t reflect all the complexity of the relational model

Import XML Templates

PLANNING 2010 « Spectrodatabase »

• Part of the VAMDC VO, Registered as service• Necessity of a mapping to VAMDC dictionary keywords• Interoperability layer to support the TAP protrocol

– Real-time request with ACL acts as a table-view• Query language supported : VAMDC SQL Subset 1 & 2 (VSS)• Output language:XSAMS

– http://www.vamdc.org/documents/standards/• Future evolution : Integration with other VOs

VO / VAMDC