Transcript PowerPoint Presentation - Chinese Virtual Observatory

Chinese Virtual Observatory
Searching for Tidal Streams in
SDSS
刘超
中国科学院国家天文台
Why search for Tidal Stream
• Galactic Structure
– Shape, Kinematics, chemical properties, etc.
• Galactic Halo Origin
– Two models debate
• Cold Dark Matter Model
– More dwarf galaxies disrupted
China-VO 2006, Guilin
11/29-12/03
2
Currently Known Tidal Stream
• Sgr dSph Tidal
Streams(Majewski03;Belokurov06b)
• Virgo Stream(Juric05)
• Monoceros Ring(Newberg02)
• Orphan Stream(GD06b)
• GD-1(GD06b)
• NGC5466 Tidal
Tail(Belokurov06a)
• Pal 5 Tidal Stream(GD06a)
• NGC5053(Lauchner06)
• NGC4147??
China-VO 2006, Guilin
11/29-12/03
3
UMa I
CVn I
Boo
Will 1
Com
UMa II
CVn II
Segue 1
References:
Willman05a Her
Willman05b
Zucker06
Zucker06
Leo IV
Belokurov 06c
China-VO 2006, Guilin
Belokurov06d
11/29-12/03
4
Why Use SDSS
• Study methods
–
–
–
–
Star count analysis
Kinematics
Chemical composition
Comparison with other galaxies
• Star count analysis is a prompt way
• Large sky area survey
– DSS
– 2MASS
– SDSS
• SDSS
– Deep space and mass dataset
– Precise photometry
– Cover Galactic North Pole
China-VO 2006, Guilin
11/29-12/03
5
Our Approach
• Binning a wide area sky
– RA=120~270deg, DEC=25~70deg
– i=19~22mag, g-i=0~1mag
– Step=0.05deg
• Pick out all over-densities
– 2sigma higher above background
• Color-Magnitude feature analysis
– Isochrone line matching
China-VO 2006, Guilin
11/29-12/03
6
Results
China-VO 2006, Guilin
11/29-12/03
7
China-VO 2006, Guilin
11/29-12/03
8
GD-1
Our Result
GNP
NGC5466
Orphan
Monoceros
China-VO 2006, Guilin
11/29-12/03
9
Conclusion
• Ten over-densities are most likely dwarf
spheroidal galaxies or star clumps on
tidal streams
• Nine over-densities and Four known
satellites compose a remarkable arc
– Possibly a tidal stream
• Distances are likely related to metalicity
for the over-densities and known dSphs
China-VO 2006, Guilin
11/29-12/03
10
Chinese Virtual Observatory
China-VO Data Access Service (DAS)
刘超
中国科学院国家天文台
China-VO 2006, Guilin
11/29-12/03
11
Goals
• Access mass data
–
–
–
–
Query data from all over the world
You need a BIG hard disk when study SDSS data
database knowledge is necessary
Furthermore, a lot of time are spent on data r/w:
download data, save temp data, format transformation
– Manage your data by yourself
• DAS goals are simply do all above for you
–
–
–
–
Let you focus on science and algorithms
Save your query time and disk space
Simplify data transferring and format transformation
Manage your data on line
China-VO 2006, Guilin
11/29-12/03
12
Functions
• Multiple ways to access data
– By a client application
– By a web page
– By web service interface
• Image, Spectroscopy data as well as Catalog data
• Data query result transfer
– FTP, GridFTP, etc.
• Data query result format transformation
– ASCII, VOTable, FITS, etc.
• Cross match among distributed catalogs
• Function scalability
– Add new databases
– Add data mining tools
– As a atomic service in a workflow
China-VO 2006, Guilin
11/29-12/03
13
China-VO DAS
Architecture
DAS WSRF Service Interface
Task Queue
ADQL Parser
Execution Plan
DataResource
WorkThread
Metadata
• DAS Server
– A grid service provider
complies with WSRF
DAS Log
DataResource Map
Sessions
Registry Proxy
OGSA-DAI Client
Authorization
• Data Node
MySpace Client
Invoke/Return
Data Transfer
– A stand alone java
application
– A series of web page
– A program coded by
users
MySpace Service
Register
Registry
Upload/Download
• Client
Query
– An OGSA-DAI server
provides multiple data
resources to community
Upload/Download
GT4 Java WS Core
Data Node
OGSA-DAI Service
Activities
Data Transform
XMatch
Image Query
Spectrum Query
Catalog Query
Data Delivery
MySpace Client
CompuCell*
Data Resources
China-VO 2006, Guilin
Authorization
Metadata
11/29-12/03
GT4 Java WS Core
14
Features
• ADQL for all
– Catalogs, Images and Spectroscopies
• Asynchronous query for mass data
• Not a system but a community
– Anybody can publish their database as a Data Node and
share them to all users
• Users can combine data query into their programs by Grid
Service Interface so that data need not to be downloaded
to local disk and data format will not be a problem
• An unified entrance for all kinds of astronomical data
• Basis of data mining tools
– Send computation to data server in future
China-VO 2006, Guilin
11/29-12/03
15
More details
• DAS server: Tian Haijun
• Cross match & ADQL execute planning:
Gao Dan
• Discover Data Node & Multi-type data
support: Lu Yong
• Client & Data Node: Yang Yang
China-VO 2006, Guilin
11/29-12/03
16
Actions
• Submit a query
• Asynchronous execution
• Data federation in distributed
environment
• Data format transformation
• Data transportation
• Job tracking
China-VO 2006, Guilin
11/29-12/03
17
Status & Future work
• Status
–
–
–
–
DAS server can run a simple job without Data Node
A java application Client is in developing
A Data Node test server is established
Data Node Discovery is in developing
–
–
–
–
–
–
–
The system query data in next spring
Multiple data format support
Distributed cross match
Connect with MySpace?
Data mining tools integration (e.g. JDL)
Visualization Integration
LAMOST data server?
• Future work
China-VO 2006, Guilin
11/29-12/03
18
Thanks!