Galactic 2

Tutors: Caroline BOT, Enrique SOLANO, Sebastien DERRIERE

Participants

Alvarez-Iglesias (Linux), Hacar (Linux), Kovacevic (Windows), Oreiro (Linux), Sanchez Ayaso (Linux), Santander-Vela (Mac OS X), Valdivielso (Linux/Windows), Vinatier (Windows/Linux), Zloczewski (Linux)

Access to tools

• VOSA register on the website prior to using VOSA
• Aladin you can download the latest official version
• TOPCAT download the latest version or start it with WebStart

Use Case 1

Based on the case presented by Alvarez Iglesias

VO-Tools: VOSA, TOPCAT

Goal:

We propose to search for L and T candidate members in very young clusters (<10Myr) using data from UKIRT Deep Sky Survey (UKIDSS) and IRAC/Spitzer and a new method based on their Spectral Energy Distribution (SED) characterization.

Background:

The knowledge of the complete census of substellar objects in star forming regions with masses below the Deuterium burning limit (M=13 M_Jupiter, Saumon et al. 1996), nicknamed isolated planetary-mass objects (IPMOs) is very important to understand the formation processes of such low mass bodies. The origin of both IPMOs and PMOs around stars remains uncertain. It is likely that IPMOs form as a natural extension of the process that leads to the formation of low-mass stars and, probably, brown dwarfs, but they could also originate in proto-planetary discs and be ejected through dynamical interactions (e.g. Reipurth & Clarke 2001). One example of the usefulness of these searches is the deep photometric study by Caballero et al (2007) in the young (~3 Myr) and relatively close (~350 pc) Sigma Orionis cluster. In this case, they combined optical, near- and mid-IR data to characterize the sub-stellar population of the cluster down to ~ 5 M_J (~M6-L6) in an area of 790 arcmin^2. Their study indicates that both brown dwarfs and IPMOs could form as an extension of the low-mass star formation process.

An optical, near-infrared and mid-infrared photometric study on field M to T dwarfs is shown in Patten et al (2006). In this case, photometry from the Sloan Digital Sky Survey and IRAC/Spitzer was used to analyze the colors of 86 sources. They show that single colors are insufficient to determine univocally the spectral type. Spectral typing accuracy increases with the number of colors used in the analysis, which is virtually the same as having a well defined SED for each possible candidate. The J-H and H-K colors for T dwarfs become bluer with increasing spectral subtype, becoming degenerate with the colors of higher mass K and M dwarfs (Patten et al. 2006). IRAC/Spizer data are fundamental to break the degeneracy amongst different types shown in the near-IR. The four IRAC channels include the CH4 fundamental absorption at 3.3 microns, the continuum peak that is present for all objects cooler than 3000 K and the CO absorption at 4.7 microns, the H2O absorption in the 5.8 micron band and CH4 absorption in the 8 micron band.

We propose use the VO capabilities to perform a search for late M, L and T dwarf members in Orion based on the characterization of their SEDs. They will be constructed using data from UKIRT Deep Sky Survey (UKIDSS) and IRAC/Spitzer. We propose to compare the SED of all objects in the studied area with template observed SED’s of late L and T dwarfs using a least-squares minimization method similar to that proposed by the Bayo et al. 2008. They developed a tool (VOSA) within the VO to analyze and compare SEDs of sub-stellar objects with theoretical models. A similar study to Caballero et al. (2007) but in a more extended area could be done in the frame of the VO-AIDA 2009 School. The use of the SED-fitting method for detection of L and T’s using VO tools could be tested directly in the Sigma Orionis cluster, since despite the detailed work by Caballero et al., the full population of IPMOs is not known to its completeness. The method can also be tested on a different young cluster in Orion, like Epsilon Orionis or Delta Orionis. We expect that with the SED-minimization technique we can improve the detection efficiency of IPMOs in young clusters and find some new candidates that may remain uncovered from color-based searches. This scientific case will fully exploit the goals of the VO-AIDA 2009 School, since it will require amongst others the use of catalogue and table handling, image and spectra handling and least-squares minimization of large amounts of data. This project will need to make full use of ALADIN, TOPCAT and VOSA amongst other VO tools. For the spectral-fitting step, we would ideally use VOSA to fit to template observed spectra, although good results can also be obtained by fitting to model spectra.

Workflow

• Spitzer / IRAC photometry is provided by the user in four files (one for each IRAc channel). Spitzer_IRAC_1.txt (3995 rows), Spitzer_IRAC_2.txt (4313 rows), Spitzer_IRAC_3.txt (5988 rows), Spitzer_IRAC_4.txt(5507 rows).

• To select the objects with photometry in the four IRAC channels we will use TOPCAT. We recomment to use the Standalone Jar file.

• • Use File/Load Table to upload the files. Use Format=CSV
  • The Scatter Plot option (number 10 in the Menu starting from the left) is used to visualize the IRAC fields. Three of them can be seen here.
  • Objects in the four IRAC channels can be found using the Quadruple Match option available from the Joins Menu in the Control Menu. Using a maximum error of 0.3 we get 22 objects. (All_spitzer.txt)

• The Spectral Enery Distribution of the selected objects is built using VOSA

• • The file All_spitzer.txt is adapted to the format expected by VOSA (All_spitzer_vosa.txt)
  • You can use a little application to make this transformation: text2vosa.php
  • The file is uploaded in VOSA using the Upload file option
  • IRAC photometry can be complemented with VO photometry. For the school select only 2MASS (radius=2 arcsec) and UKIDSS/Galactic Cluster Survey photometry).The Sigma Orionis region has not been observed by SDSS. NOTE: Sometimes data downloading from UKIDSS takes time. If this is the case, repeat the search selecting 2MASS photometry only.
  • The physical parameters are obtained by fitting the observational SEDs with different collections of theoretical models. Select NextGen , COND and DUSTY.

Use Case 2: Discovery of Brown Dwarfs mining the 2MASS and SDSS databases

VO-Tools: ALADIN, TOPCAT, VIZIER

Background: Brown dwarfs are objects occupying the gap between the least massive stars and the most massive planets. They are intrinsically faint objects so their detection is not straighforward and, in fact, was almost impossible until the advent of global surveys at deep optical and near-infrared bands like SDSS, 2MASS or DENIS among others.

We propose here to mine the SDSS and 2MASS databases to identify T-type brown dwarfs through an appropriate combination of colours in the optical and the infra-red, an approach that perfectly fits into the Virtual Observatory.

Workflow:

With this use case, we explore different ways to do the same tasks (cross-match, sources selection,...) with different VO tools. Given the capabilities of each VO tool, there can be indeed several ways to execute the same workflow (series of steps in the data processing):

• First flow, with Aladin: ALADIN is used to
  • Search 2MASS and SDSS sources around RA:08h 30m DEC:01d 30m (default radius: 14 arcminutes).
    • 2MASS: 686 sources
    • SDSS: 9855 sources
  • Find common sources in 2MASS and SDSS catalogues (use the Aladin catalogue cross match tool with default threshold (4 arcsec) and bestmatch option): 680 sources (677 point sources).
  • Select points sources using the SDSS flag (cl=6) creating a filter plane (you need to edit a filter in advanced mode to perform this selection - see the Filter manual for detailed help): 642 sources (639 point sources)
  • Create a new plane with the filtered sources
  • Select sources with no detection in the u,g SDSS filters (u > 22.0 & g >22.2) using a new filter plane: 6 sources
  • Select sources following the criteria provided by Burgasser et al. (2000, Apj, 531, L57).
    • (J-H)<0.3 && (H-K)<0.3 : 1 object --> RA:127.703265deg; DEC:1.475320deg
    • To do so, you can either use arithmetic operations between columns directly in the filter syntax, or you can create new columns in the x-match catalog that are an arithmetic combination of 2 columns (e.g. Jmag & Hmag)
  • Confirmation of the brown dwarf nature of this object by searching through VO services using the Load from the Virtual Observatory option.

• Second flow: Repeat the same analysis using TOPCAT.
  • Start Aladin and TOPCAT tools
  • Search the 2MASS and SDSS sources with a cone search from TOPCAT around RA:08h 30m DEC:01d 30m
    • Go to the File menu and "Load Table"
    • Go to the DataSources menu and select "Cone Search"
    • Give "2MASS" as a Keyword to query available services and select the 2MASS point sources catalog
    • Give the coordinates and a radius of 14'
    • Repeat for the SDSS search (select the DR6 SDSS calatog for example)
    • Alternatively, you could load the 2MASS and SDSS catalogs in Aladin and broadcast the planes to TOPCAT using SAMP.
  • Create a new table by matching raws in the two tables with 4" maximum error and the best match option. 680 pairs are found.
  • Visualise the cross-matched raws and define a raw subset for which the SDSS class is equal to 6
  • Select sources with no detection in the u,g SDSS filters (u > 22.0 & g >22.2) by defining a new raw subset (6 remaining sources)
  • Update the selection criteria to add the criteria:
    • (J-H)<0.3 && (H-K)<0.3 : 1 object --> RA:127.703265deg; DEC:1.475320deg
    • As in the previous flow, you can either give selection criteria that are an arithmetic combination of columns, or you can create new columns in the catalog. For this, go to the main TOPCAT window -> Display column meta data -> Columns menu -> New synthetic column
  • To visualise the information on the brown dwarf candidate, you can select the name of your subset in the "Row Subset" menu in the main TOPCAT window and broadcast it to Aladin (if Aladin is opened) or any other VO application compatible with the SAMP interoperability standard

• Third flow: Do the query and selection with Vizier, cross-match with TOPCAT or Aladin
  • Vizier enables you to query catalogs with selection criteria. Go to the Vizier webpage http://vizier.u-strasbg.fr/viz-bin/VizieR
  • Type "SDSS" in the description, give the RA:08h 30m and DEC:01d 30m coordinates as well as target radius of 14 arcminutes, and press "Find Catalogues"
  • Select the SDSS Photometric Catalogue
  • Give constraints on the class of objects (=6), the u magnitude (>22.0) and g magnitude (>22.2) columns
  • Change the number of entries per table to a large number (in order to see the 5829 sources) and the output to VOTable (doing so, the table will be readable by all VO tools).
  • Read the VO table file with Aladin or Topcat and proceed as described above (first flow for Aladin, second flow for TOPCAT)

• Advanced scripting in Aladin: Aladin has a script mode, where you can build a list of commands to be processed (Tool > Macro Controller). The workflow can be executed automatically for a list of targets.
  • Open the Tool > Macro Controller in Aladin
  • Load the script (Aladin_workflow_script.ajs)
  • Load the parameters (Aladin_workflow_params.txt)
  • Execute the script for one or all the parameters
  • You can try to modify the script to save some results for example (Help > Help on script commands)

• Aladin - Topcat interaction: We have extracted from SIMBAD a list of 1108 objects flagged with an object type "brown dwarf": (sim-sam_bd.xml) in VOTable format. We will see how we can use the interaction between Aladin and TOPCAT using SAMP to perform analysis on this file.
  • Make sure you have Aladin and TOPCAT running (run VODesktop first, or run Aladin, then topcat if you have Aladin v5.019)
  • Load this VOTable file in Aladin (or TOPCAT)
  • If you want an all-sky background image, use the allsky.fits image, or use the hidden feature in Aladin : (Ctrl+i) or File > Load astronomical image > Aladin image server, and type "allsky" as Target
  • Broadcast it to the other application with SAMP (or PLASTIC)
  • Display a J-H vs H-K color-color diagram in TOPCAT
  • Different object types (M, L or T dwarfs) are located in different regions of the diagram, as described in this study

• • Select objects (by defining a subset in TOPCAT and broadcasting it to Aladin) corresponding to the L dwarfs locus (green circles), the T dwarfs (red stars) and M dwarfs (cyan triangles), and see in the Aladin measurement panel what Sp_Type they have in SIMBAD.
  • Conversely, you can also:
    • select all simbad objects in Aladin (double click the plane or use right-click > select all)
    • detach the Aladin measurements ( icon above measurements)
    • sort them by Sp_Type (click on table header)
    • hover the mouse on measurement lines
    • see that the corresponding object in the TOPCAT color-color diagram are highlighted
    • you can have more interactions by defining/broadcasting subsets in each tool

Use Case 3

Proposed by: Luka C. Popovic, Dragana Ilic and Andjelka Kovacevic

Obtain as big as possible sample of AGN type 1 (broad emission line present with w/λ>1000km/s) spectra that contain Balmer line series (Hα to Hε) and helium lines He II 4686 and He I 5876 with S/N=20.

--> This is not a galactic case

Use Case 4

Proposed by: Bojan Arbutina, Dragana Ilic and Andjelka Kovacevic

Observations in narrow band filters (wavelength at Hα, [SII], and continuum ~6500) of Holmberg IX galaxy (Equ J2000.0 09h57m32.0s +69d02m45s)

--> No observation in those bands.

Feedback from participants

Pros

The detailed instructions in each science cases, which permits to have a first use of VO-tools alone even if we never saw these tools before.

Good opportunity to see different applications of using VO tools and to learn how to apply them to your own research. I liked the fact that most of the emphasis of the school has been put on the students really using the tools, rather than attending to talks.

The large number of tutors and their disponibility to answer all questions.

Excellent tutors (communication,provided information, answers) and group working dynamics

Cons

Perhaps the only drawback that I see is that it has been a lot of information to be able to process in only a few days. There was not enough time to try to put the hands on other cases appart from the ones in your own group. This has limited us somehow to use only 3 or 4 tools. It would have been nicer to be able to try some other tools.

Improve availability of asrogrid registry in topcat cone search.

Remarks

-- PaoloPadovani - 02 Mar 2009

• VOsoftwarecomments.txt: Feedback about software (Kamil Zloczewski)