science with the vo mark allen (cds, observatoire de strasbourg)
DESCRIPTION
Example science drivers Find all info on a given set of objects Defined by positions, colours, morphology,... Build SEDs from multi-archive data Accounting for instrumental, sensitivity, aperture effects “Outlier Science” Multi-d parameter searches Compare LSS with 'virtually observed' N-body simulations Re-analyse the SLOAN, MACHO Multi- census of AGN Build a survey to search for Cosmic ShearTRANSCRIPT
Science with the VO
Mark Allen (CDS, Observatoire de Strasbourg)
Outline Example Science Drivers Science 'Reference Mission' The 'VO approach' & Common Aspects Demonstrations and Early VO Science Published examples Challenges and Opportunities
Example science drivers Find all info on a given set of objects
Defined by positions, colours, morphology,... Build SEDs from multi-archive data
Accounting for instrumental, sensitivity, aperture effects
“Outlier Science” Multi-d parameter searches
Compare LSS with 'virtually observed' N-body simulations
Re-analyse the SLOAN, MACHO Multi-census of AGN Build a survey to search for Cosmic
Shear
AVO Science Reference Mission
Key scientific results that a European VO should be able to achieve when fully implemented
Consists of science cases over broad range of astronomy and the related requirements
AVO Science Advisory committee (2005)http://www.euro-vo.org/pub/fc/cases/srm.pdff
Science Cases Circumstellar disks: from pre-Main Sequence stars to stars
harbouring planets Intermediate Velocity Clouds Which Star will go Supernova next? Initial Mass Function within 1kpc: Planetary to Stellar Masses Initial Mass Function for Massive Stars Contributions of Low and Intermediate Mass Stars to the ISM Galaxy Formation and Evolution from z=10 to 0.1 Build-up of Supermassive Black Holes Formation and Evolution of Galaxy Clusters Correlation of CMB, radio/mm and optical/NIR Galaxy Surveys
Galactic
Extragalactic
Cosmological
Galaxy Formation and Evolution from z=10 to 0.1
Required data Deep Multi-wave surveys (GOODS, COSMOS) HST+ACS bviz imaging SLOAN Optical spectroscopy MERLIN, GMRT, VLA, ATCA radio Chandra and XMM-Newton X-ray Spitzer mid-IR Future sub-mm GALEX UV imaging
When did the 1st objects form? What are the progenitors of present day massive ellipticals? How many massive galaxies at z>1,2,4? How do SF and galaxy stellar mass densities evolve?
“Only now, and only with through the VO, are the datasets large enough, and the tools mature enough that Galaxy Formation and Evolution can be examined in a meaningful way.”
VO approach to the problem Extract sample from data
Perform Sextractor type photometry Cross correlate with images, catalogues, spectra.
Crucial that output results are scientifically useable and reliable Matching of PSF, consisitent photometric apertures, treatment of noise, and upper limits
Sanity checks like stellar colours Output multi-band catalogue, and colour-colour diagrams
Visualize output colour-colour space Photometric z from SEDs (Template SED libraries, extinction curves etc.) Physical Parameters – L, E(B-V), SFR, M/L stellar mass
Comparison with star formation scenarios and synthetic spectra Morphological analysis
Stack images at same wavelength, or spectra at different redshifts
Build average spectra for specific object classes Angular clustering analysis
Comparison with mock catalogues from theoretical simulations
Common aspects Browsing/searching of data and
distributed information Manage large amounts of distributed
heterogeneous data Combining Multi-wavelength data taking into
account different: Units, photometric systems spatial, wavelength and time coverage resoultions/PSF, observing technique
Cross-Matching Catalogues: Large-Small, Sparse-Dense Taking into account:
positional uncertainties, resolution, completeness extra constraints: e.g. colour, object type, environment
Compare observations with models Virtual observations of models Projection of models to observed parameter
space Spectral fitting/classification
Science Demos and early VO Science
Exploit archives as virtual telescopes (3 proposal cycles 2000-2003)
VO science on HST, ESO: VLT, NTT, WFI, ISO archives
advanced searching of achive exposure/seeing/field density location e.g. for pre-discovery images for asteroids
http://www.euro-vo.org/astrovirtel/index.html 20+ publications listed on AstroVirtel pages
Early demonstration project identifies new L-type Brown Dwarf
Found in SDSS/2MASS correlation designed to reproduce known (~200) Brown Dwarfs
SDSS 15M sources, 2MASS 160M Was done as protoype, now used as NVO
OpenSkyQuery training example
Demo 2004: First Science Searching for type 2 Quasars using Virtual
Observatory tools
Demo 2005 AGB to Plantary Nebulae transition
Workflow: SEDs to Models
Demo 2003: First Light VO Visualisation and Data Discovery
- GOODS Optical ACS data & catalogues - Chandra X-ray catalogues - Select absorbed X-ray sources - Cross-match X-ray and optical - Apply empirical estimator for Lx - Check against spectroscopy Lx>1044 : QSO 2
Discovery of QSO 2s with VO tools
Science Conclusions Using
the deepest Chandra X-ray, and HST imaging
& Empirical estimator for Lx we find 68 New type 2 AGN 31 Qualify as QSO 2, z~4 Many more QSOs than predicted
Published: Padovani, Allen, Rosati, Walton A&A 2004
More published results
• Using VO tools to investigate distant radio starburst hosting obscured AGN in the HDF(N) region, Richards et al., A&A, 472, 805 (2007)
• Albus 1: a very bright White Dwarf candidate, Caballero & Solano, ApJ, 665, L151 (2007)
• Flare productivity of newly-emerged paired and isolated solar active regions, Dalla, Fletcher, & Walton A&A, 468, 1103 (2007)
• Radio-loud Narrow-Line Type 1 Quasars, Komossa et al., AJ, 132, 531 (2006)
• Luminous AGB stars in nearby galaxies. A study using VO tools, Tsalmantza et al., A&A, 447, 89 (2006)
• + more...
Example: Serendipitous Outlier
Caballero & Solano 2007, ApJ, 665, L151 'ALBUS 1: A very bright white dwarf candidate'
found a bright source with very blue VT-Ks colour appreciable proper motion mag. and colours --> White Dwarf
Candidate
Found in optical-NIR study to characterize young stars and brown dwarfs surrounding εOri and δOri
VO aspects: Correlate Tycho-2 and 2MASS catalogs
[in 10 45' radius fields at same Gal. lat.] also: SuperCOSMOS, USNO-B1, NOMAD
and DENIS excluded radio(VLA), mid-IR(IRAS), UV
(EUVE), X-ray(ROSAT)... or object in literature (SIMBAD)
Aladin & VO catalog interoperability
spectrum obtained: ALBUS 1 is a Helium-rich subdwarf Also known as CPD -20 1123 (Gill & Kapteyn
1896) (and in VizieR catalogs...)
published Aug 2007, and then ...
Example: Solar Physics2 papers utilizing AstroGrid workflow
system1. 'Flare productivity of newly-emerged
paired and isolated solar active regions' Dalla et al., 2007, A&A, 468, 1103
2. 'Invisible subspots and rate of magnetic flux emergence' Dalla et al. 2008, A&A, 479, L1
Analysed 7000+ sunspots from over 25 year period
Some areas of the disk are missing new sunspots compared to others
because when they are gowing rotation caries them away from centre of disk
AstroGrid workflows
We used AstroGrid workflows to:
(1) retrieve region catalogues and identify ‘new’ regions;
(2) study the location of emergence with respect to pre-existing regions; and
(3) querying catalogues of flares to establish flare productivity.
Visualisation of results using Topcat.
Slide courtesy of S. Dalla
Fossil Groups in the SDSS US NVO Summer School The National Virtual Observatory:
Tools and Techniques for Astronomical Research. ASP vol. 382
2007 AJ, 134, 1551 Santos, Mendes de Oliveira & Laerte Sodre
Fossil Groups – systems with M and LX comparable to clusters of galaxies, but dominated by single isolated large elliptical galaxy
'Fossil' – comes from formation scenario:collapsed early -> oldest, undisturbed
systems only 15 fossil groups known in literature
VO methods to identify new fossil groups in SDSS DR5 – increasing known by x3
Use of SQL - Structured Query Language ADQL – Anstronomical Data Query
Lang. OpenSkyQuery VO service
Steps Elliptical Galaxy Selection
luminous red galaxies r<19 – 112510 sources Cross-Match with X-ray data
OpenSkyQuery xmatch of SDSS with RASS constrained by positional uncertainties
Selection of Companions spectro. and photom. redshifts
Photometric Condition for Fossil Groups mi > m1 + 2
Analysis of dominant galaxy Cross-check against known systems
Simple expressions applied to large data set in powerful way
-> New Fossil Groups Query Language
expressions published in the paper
My observations of current VO science
largely about getting and combining data
once data obtained/selected analysis is mostly by traditional methods
Workflow systems ramping up (esp. AstroGrid) Scripting with familiar lang. (Python) very
promising End-to-End with VO ?
in practice VO is being used where it is needed more and more papers acknowledging VO
Where's the detailed scientific interoperability?
as required for SRM like problems
first IVOA standards -> good data connectivity
in progress: characterisation of data – data models units, PSFs, time/spatial/energy coverage theory and modeling... (not to mention Grid, VOSpace, Curation, Access
Layers)
Comparing with Theory Simplistic demos
show interoperability but immaturity
Need for meetings like this!
OpportunitiesEuro-VO AIDA Calls for Proposals to carry out Astronomy projects
that would benefit from VO. Support from Euro-VO scientists
Watch for announcements!