Transcript sources Cat1 Cat2

THE US NATIONAL VIRTUAL OBSERVATORY
VIM
Visual Integration and
Mining
A Tool to Explore Your Sources
Roy Williams
Caltech
us-vo.org/vim
Start with your sources
(currently <1000)
• Multiple named sky positions
– Simple text (RA,Dec,ID)
– Text wizard (NVO table importer, today 17.10)
– VOTable of positions (local or url)
– Push positions from NVO portal
us-vo.org/vim
Getting Data about
your sources
us-vo.org/vim
Cen A @ 840 MHz
Skyview cutouts in many wavelengths
us-vo.org/vim
hover mouse to
enlarge
Skyview cutouts in many wavelengths
Abell clusters in
optical and X-ray
us-vo.org/vim
Spectra from VO SSAP services
Hover mouse on
thumbnail to see
larger image
Spectral Collections from JHU
brokered by NRAO:
• Arecibo Maser Catalog
• 2dFGRS
• SDSS DR5
us-vo.org/vim
Getting Data about
your sources
Proximity search
us-vo.org/vim
Proximity Search =
Multicone
user-supplied
sources + radius
returns VOTable
radius
us-vo.org/vim
VIM tables structure
– Multiple matches per source
– Can join to sources (outer or nearest)
sources
Cat1
RA Dec ID
us-vo.org/vim
Cat2
Shakbazian galaxy groups
prox search in Simbad
us-vo.org/vim
Table column metadata
NED
metadata
us-vo.org/vim
us-vo.org/vim
Choose which
columns to
display
Compute new
columns, eg joint
IR-optical colors
us-vo.org/vim
Visualization
by download
& use Topcat
us-vo.org/vim
Architecture
All the relevant information about your sources
-- mashups from the VO
-- kept for you in persistent storage
-- view, mine, download
NVO portal
Registry
Inventory
browser
upload sources
HTML + AJAX
Catalogs (cone)
Images (Skyview)
Spectra (SSAP)
Vim
python api
personal
persistent
storage
us-vo.org/vim
• Auto-script:
• Clicking builds script
• Can edit script and rerun
us-vo.org/vim
Vim auto-scripting
# got some sdss sources from Vizier, a 12 minute radius with rmag<21
# about 773 sources found
sources.run(bench, open(‘mysources.votable').read())
# run 2mass prox search at 2 arc seconds
proxsearch.run(bench, survey='2MASS-PSC', sr=0.00055556, tablename='_2mass', verb='1', maxthreads='10')
# merge with the sdss
join.run(bench, tablename='_2mass', match='best')
# select rows with non-null j magnitude (i.e. thos in 2mass)
# we get abouyt 148 remaining
selectrows.run(bench, tablename='sources', criterion='! NULL__2mass_j_m')
# compute 2mass-J - SDSS-r
addcol.run(bench, tablename='sources', colname='JminusR', expression='_2mass_j_m - rmag')
# compute 2mass-H - SDSS-z
addcol.run(bench, tablename='sources', colname='Hminusz', expression='_2mass_h_m - zmag')
# then download sources.votable from Utility/ViewAllFiles, and plot with topcat
us-vo.org/vim
VIM Features
• WebServer or Laptop install
• Pure python webserver (CherryPy)
• Scalable
• Auto-script, edit, rerun paradigm
• Powered by Stilts (2,000,000,000 rows !)
• Simple sharing
• Share your workbench by emailing a URL
http://envoy5.cacr.caltech.edu:8888/?benchID=7462948574635185948473627485958433
• Can write-protect with password
us-vo.org/vim
VIM Features
• Content
–
–
–
–
Any cone search (all 1000’s in VOregistry)
Cutouts from Skyview
NVO Inventory services
Spectra via SSAP (from NRAO)
• Thumbnails and images and FITS
• Display
–
–
–
–
Column selection, Row sort/select
Images small-hover-large
Tools and metadata choice
AJAX progress for data fetching
…but not Safari :-(
us-vo.org/vim
NVO Portal:
Using the registry to
get catalogs
us-vo.org/vim
NVO Portal:
Finding catalogs by
spatial correlation
us-vo.org/vim
NVO Portal:
Inventory service
Catalog A:
Perfect
correlation
Catalog B:
Weak
correlation
my sources
Catalog C:
No correlation
(1000’s of catalogs ingested)
us-vo.org/vim
Questions?
us-vo.org/vim