Docstoc

Handling 3D Data in the Virtual Observatory

Document Sample
Handling 3D Data in the Virtual Observatory Powered By Docstoc
					ADASS XV – San Lorenzo de El Escorial – 4 Oct 2005

Handling 3D Data in the Virtual Observatory
Igor Chilingarian (CRAL Observatoire de Lyon, France/SAI MSU, Russia) Francois Bonnarel (CDS Observatoire de Strasbourg, France) Mireille Louys (CDS Observatoire de Strasbourg, France) Jonathan McDowell (Harvard-Smithsonian CfA, USA)

What is 3D spectroscopy (1)?
IFU Spectroscopy

I. Chilingarian et al. - ADASS XV - San Lorenzo de El Escorial, 4.Oct.2005

What is 3D spectroscopy (2)?
Scanning Fabry-Perot Interferometer

Data Processing

Phase Surface

I. Chilingarian et al. - ADASS XV - San Lorenzo de El Escorial, 4.Oct.2005

Storing 3D Data in FITS



Pure 3D data cube (for IFP data and for some IFU) 2D-image (one spectrum per row) + binary table  Euro3D Format

Euro3D Format •FITS binary data table: one row per spectrum •Binary table describing shape of spatial elements (“spaxels”) •Some mandatory metadata, including: common spectral WCS for all spectra, common spatial WCS for all spatial elements, meteo parameters during the observations, etc.

I. Chilingarian et al. - ADASS XV - San Lorenzo de El Escorial, 4.Oct.2005

Characterisation DM
The basic part of the most general data model: Observation DM Provides a physical characterisation of a dataset

Level 1 Level 2 (pos) Spatial

Coverage Coverage Location

Resolution Resolution Ref.Value

Sampling Sampling Ref.Value

Temporal (time) Coverage Location Ref.Value Sampling Ref.Value Level 3 (pos) Resolution Spatial Bounds Bounds Bounds Spectral (em) Location Ref.Value Ref.Value Temporal (time)Coverage Bounds Bounds Sampling Bounds Level 4 (pos) Resolution Spatial Support Support Support Observable Location Ref.Value Ref.Value Spectral (em) (phot) Bounds Bounds Bounds Temporal Support Support Support Spatial (pos)(time) Map Map Map Observable Bounds Bounds Bounds Spectral (em) (phot) Support Support Support Temporal (time) Map Map Map Observable Support Support Support Spectral (em) (phot) Map Map Map Observable (phot) Map Map Map

I. Chilingarian et al. - ADASS XV - San Lorenzo de El Escorial, 4.Oct.2005

Characterisation: relation to STC

I. Chilingarian et al. - ADASS XV - San Lorenzo de El Escorial, 4.Oct.2005

Characterising IFU datasets (1)
Characterization[ucd=time]/Coverage/Location/coord/Time/Value Only first two levels (Location/Ref.Value and Bounds) is usually the only temporal-axis-related information preserved by the data should be pipelines for the whole dataset, because further processing provided levels become too difficult for understanding. The rest should be done for each individual IFU spectrum
Characterization[ucd=pos]/Coverage/Location/coord/Position2D/Value2 Characterization[ucd=pos]/Resolution/ReferenceValue Characterization[ucd=pos]/Sampling/ReferenceVal ue

Characterization[ucd=pos]/Coverage/Bounds/limits/LoLimit2Vec Characterization[ucd=pos]/Coverage/Bounds/limits/HiLimit2Vec

I. Chilingarian et al. - ADASS XV - San Lorenzo de El Escorial, 4.Oct.2005

Characterising IFU datasets (2)

I. Chilingarian et al. - ADASS XV - San Lorenzo de El Escorial, 4.Oct.2005

Characterising IFU datasets (2)
Characterization[ucd=phot]/Coverage/Location/coord/Flux/Value

Characterization[ucd=em]/Coverage/Location/coord/Spectral/Value

Characterization[ucd=em]/Resolution/ReferenceValue  mean spectral resolution (FWHM) Characterization[ucd=em]/Sampling/ReferenceValue  mean sampling (usually constant) Characterization[ucd=phot]/Resolution/ReferenceValue  1 e- (for CCD) Characterization[ucd=phot]/Sampling/ReferenceValue  1 ADU (for CCD)
I. Chilingarian et al. - ADASS XV - San Lorenzo de El Escorial, 4.Oct.2005

Characterising IFU datasets (2)

Characterization[ucd=phot]/Coverage/Bounds/limits/LoLimit Characterization[ucd=phot]/Coverage/Bounds/limits/HiLimit

Characterization[ucd=em]/Coverage/Bounds/limits/LoLimit Characterization[ucd=em]/Coverage/Bounds/limits/HiLimit Characterization[ucd=em]/Resolution/Bounds/limits can be computed using special technics Characterization[ucd=em]/Sampling/Bounds/limits are not defined Characterization[ucd=phot]/Resolution/Bounds/limits are [1e-,1e-] for CCD Characterization[ucd=phot]/Sampling/Bounds/limits are [1e-,1e-] for CCD
I. Chilingarian et al. - ADASS XV - San Lorenzo de El Escorial, 4.Oct.2005

“Live” example: MPFS dataset

I. Chilingarian et al. - ADASS XV - San Lorenzo de El Escorial, 4.Oct.2005

Accessing 3D Data
 



SSAP 0.9 specifications allow to access any type of data described with any data model Delivering the 3rd and 4th levels of characterisation metadata can be done using binary extensions: output XML document will contain pointers to binary MIME-attachments Since the 2nd level is sufficient for most of the applications, data can be stored in Euro3D format within the archive

I. Chilingarian et al. - ADASS XV - San Lorenzo de El Escorial, 4.Oct.2005

Summary
   

At present 3D data in VO can be described using Characterisaton DM Latest version of the Simple Spectral Access Protocol provides all the necessary functionality for accessing 3D data Euro3D format developed especially for IFU data can be used as a format for storing such type of data by the datacenters Evenly sampled 3D data cubes (such as IFP, Radio-cubes, etc.) can be characterised as well. But we propose to use pure-3D fits for storing them, mostly because of performance reasons

All the necessary infrastructural components exist for building VO-compliant science-ready archives of 3D data. During next few months we expect to have a couple of such resources in the Virtual Observatory.
I. Chilingarian et al. - ADASS XV - San Lorenzo de El Escorial, 4.Oct.2005


				
DOCUMENT INFO
Shared By:
Categories:
Stats:
views:40
posted:11/2/2009
language:English
pages:13