Collinder 69: From SED fitting to Age estimation using VOSA


  • Step 1.- Go to http://svo.cab.inta-csic.es/theory/vosa.
  • Step 2.- To use VOSA you need to be registered. Click on "Register" and fill in the fields (email, name and passwd).
  • Step 3.- VOSA can be used to study stellar and extragalactic data. For this use case, click on "Stars and brown dwarfs".
  • Step 4.- Cut and paste in a file the list of objects in "VOSA format" included in vosa_usecase1.txt
  • Step 5.- Upload the file in VOSA (tab Files). Give a description and do not forget to select "magnitudes" as file type. Then, click "Upload".
  • Step 6.- In the new window, click on the corresponding radio button and then on "Select".
  • Step 7.- Click on "Objects" (next tag). You will see a table with three columns: The name of our objects (first column in the input file), the coordinates provides by the user (second and thrid column of the input file) and a third column where the coordinates provided by Sesame will appear once we click on "Search for obj. coordinates". As our object identifiers are meningless (LOri001, LOri002,...) we will not use the Sesame capabilities.
  • Step8.- We skip the "Distances" and "Extinction" tags as the VO services consulted by VOSA do not provide any information for our list of objects.
  • Step9.- With the next tag "VO Phot" we can complement our "user photometry" with photometry found in a number of VO services. For this use case we select only 2MASS and CMC-14. Do not forget to click on "Save VO photometry" once the results are displayed. Once this is done, a summary table with the VO photometry (in flux units) will appear.
  • Step10.- The next tag ("SED") gives us the possibility of checking the SED before the model fitting. User data are plotted in red and VO data in green. Bad photometric points can be removed cliking on "Delete". If VOSA detects an infrared excess, the photometric points are drawn in black and are not considered in the fitting process. The user can manually overrride it and specify a new limit in the "Apply excess from" panel. Do not make any modification to what VOSA shows in this page.
  • Step11.- In the next tag ("Model Fit"), different grids of theoretical models are displayed. They cover different ranges of physical parameters. For this case, we select the "Nextgen", "Dusty","Cond" and "Kurucz" set of models. Click on "Select model params".
  • Step12.- In this window, we can refine the range of physical parameters that will be used for the fit. We will make the following assumption:
    • Nextgen: Teff: 2500-6000K; logg: 3.5-4.5
    • Dusty: Teff: 1800-2500K; logg: 3.5-4.5
    • COND: Teff: 100-1800K; logg: 3.5-4.5
    • Kurucz: Teff: 3500-6000K; logg: 3.5-4.5; met: 0
    • After this, click on "Make the fit".
  • Step13.- We can now see a summary table with the best fit results. Click on "Show graphs" to have a look at the graphics. The effective temperatures obtained after the fitting would be:
    • LOri0001: 4000K (Nextgen)
    • LOri0002: 3900K (Nextgen)
    • LOri0003: 4000K (Kurucz)
    • LOri0004: 3500K (Kurucz)
    • LOri0005: 4000K (Nextgen).
  • Step14.- Alternatively to the chi2-fitting you can perform a Bayesian fitting using the "Bayes analysis" tag. To do so, we select the same collection of models as in Step11 and click on "Select model params". Then, click "Make the fit". Do not make any restriction on the range of the physical parameters. For every collection of models and every physical parameter, a summary table with information on the model with the highest probability is shown. For each object, the information is graphically displayed by clicking on the object name (top left panel).
  • Step15.- In order to estimate ages and masses for our objects we will make use of the "HR Diag." tab. The isochrones and evolutionary tracks to be used depend on the best fit model (e.g. Nextgen isochrones and evol. tracks if Nextgen was the best model). By clicking on "See list of objects" you can see the relationship between objects and tracks/isochrones. Then, click on "Make the HR diagram".
  • Step16.- You can save different type of results (plots, VO photometry, Bayes fit, chi-2 fit,...) using the "Save Results" tag.
  • Step17.- A detailed description of how VOSA works can be found in the "Help" tag.

  • Use Case #2: Physical parameter determination of field, nearby stars.
  • Step 1.- Cut and paste in a file the list of objects in "VOSA format" included in vosa_usecase2.txt
  • Step 2.- Go to the "File" tag. Upload the file. Click on the corresponding radio button and then click "Select".
  • Step 3.-
    • Click on the "Objects" tag.
    • Retrieve the coordinates of our list of objects by cliking "Search for Obj. Coordinates".
    • Click on "Mark all: Sesame".
    • Click on "Save Obj. Coordinates".
  • Step 4.-
    • Move to the "Distances" tag.
    • Choose a search radius of 10 arcseconds.
    • Click on "Search for Obj. Distances".
    • Click on "Mark all: Hipparccos".
    • Click "Save Obj. Distances".
    • Skip the "Extinction" tag.
  • Step5.- With the next tag "VO Phot" we can complement our "user photometry" with photometry found in a number of VO services. For this use case we select only 2MASS, Tycho-2 and GALEX. Do not forget to click on "Save VO photometry" once the results are displayed. Once this is done, a summary table with the VO photometry (in flux units) will appear.
  • Step6.- The next tag ("SED") gives us the possibility of checking the SED before the model fitting. User data are plotted in red and VO data in green. Do not make any modification to what VOSA shows in this page.
  • Step7.- In the next tag ("Model Fit"), different grids of theoretical models are displayed. They cover different ranges of physical parameters. For this case, only "Nextgen" is selected. Click on "Select model params".
  • Step8.- In this window, the range of physical parameters to be used in the fit can be refined. We will make the following assumption:
    • Nextgen: Teff: 2500-10000K; logg: 3.5-4.5
    • After this, click on "Make the fit".
  • Step9.- We can now see a summary table with the best fit results. Click on "Show graphs" to have a look at the graphics. The effective temperatures obtained after the fitting would be:
    • HIP103: 6200K
    • HIP169: 4000K
    • HIP38: 5400K
    • HIP436: 4400K
    • HIP636: 8000K
  • Step10.- Alternatively to the chi2-fitting you can perform a Bayesian fitting using the "Bayes analysis" tag. To do so, we select the same collection of models as in Step7 (only Nextgen) and click on "Select model params". Then, click "Make the fit". Take the same range as in Step 8 (Teff: 2500-10000K; logg: 3.5-4.5). A summary table with information on the model wit the highest probability is shown. For each object, the information is graphically displayed by clicking on the object name (top left panel).
  • Step11.- In order to see how our objects are distributed in a HR diagram, we will make use of the "HR Diag." tab. Four of them lie outside the area covered by the isochrones/evol. tracks. You can "clean" the plot by clicking on "Unmark all" (top left in "Models" table). Change the limits in the X,Y-axis (X: 3500-8500K; Y: -1.5,1.5). Click on "Plot" again. Compare what you get with the image given below. hr.jpg


  • Use Case #3: Physical parameter determination of highly reddened stars.
  • Step 1.- Cut and paste in a file the list of objects in "VOSA format" included in vosa_usecase3
  • Step 2.- Go to the "File" tag. Upload the file. Click on the corresponding radio button and then click "Select".
  • Step 3.-
    • Click on the "Objects" tag.
    • Retrieve the coordinates of HE1136-1641 by clicking "Search for Obj. Coordinates".
    • Tick on "Sesame" for HE1136-1641.
    • Click on "Save Obj. Coordinates".
  • Step 4.-
    • Move to the "Extinction" tag (skip the "Distances" tag).
    • Click on "Search for Extinction properties".
    • Add a defaul Rv value of 3.1. for obj13 and obj15
    • Click on Mark All: "User", "Larson", "Savage", "Morales" (click four times).
    • Click on "Save Ext. properties".
  • Step5.- With the next tag "VO Phot" we can complement our "user photometry" with photometry found in a number of VO services. For this use case we select 2MASS, Tycho-2, CMC-14, Stromgren and GALEX. Do not forget to click on "Save VO photometry" once the results are displayed. Once this is done, a summary table with the VO photometry (in flux units) will appear.
  • Step6.- The next tag ("SED") gives us the possibility of checking the SED before fitting a model. User data are plotted in red and VO data in green. Do not make any modification to what VOSA shows in this page.
  • Step7.- In the next tag ("Model Fit"), different grids of theoretical models are displayed. They cover different ranges of physical parameters. For this case, we select "Nextgen", "Kurucz" and "Tlusty". Click on "Select model params". Do not change the parameter range. Then click on "Make the fit".
  • Step8.- We can now see a summary table with the best fit results. Click on "Show graphs" to have a look at the graphics. The effective temperatures obtained after the fitting would be:
    • HE1136-1641: 37500K (Tlusty)
    • obj13: 8800K (Nextgen)
    • obj15: 21000K (Kurucz)
    • obj25: 9000K (Nextgen)
    • obj26: 21000K (Tlusty)
  • Step9.-
    • Go back to the "Files" tag.
    • Upload the file vosa_usecase3b
    • Tick the corresponding radio button.
    • Click "Select"
  • Step10.- Repeat Step 3 to gather the coordinates of HE1136-1641.
  • Step11.- Skip the "Distances" and "Extinction" tags .
  • Step12.- Repeat Steps 5-8. You should obtain the following temperatures:
    • HE1136-1641: 35000K (Tlusty)
    • obj13: 5000K (Kurucz)
    • obj15: 12500K (Kurucz)
    • obj25: 5500K (Kurucz)
    • obj26: 9200K (Nextgen)
  • Step13.- Click on "Show graphs" to have a look at the graphics. Apparently the fitting is excellent in all cases. What is causing the large differences in Teff is we compare these values with those calculated in Step8? The answer is the extinction (we are now assuming no extinction) which has a strong impact on the SED shape. In the figure given below, red circles represent the observed SED of a star with E(B-V)=0.76. Blue squares represent the SED of the same star after the reddening correction. Forget about the green triangles.
no_ext.png

back to main twiki page for the school

Topic attachments
I Attachment History Action Size Date Who Comment
JPEGjpg hr.jpg r2 r1 manage 42.6 K 2011-03-17 - 21:40 UnknownUser HR
PNGpng no_ext.png r1 manage 6.4 K 2011-03-18 - 09:21 UnknownUser Effect of extinction on SED
Texttxt vosa_usecase1.txt r1 manage 3.3 K 2011-03-17 - 13:10 UnknownUser VOSA. Use Case #1. Data
Texttxt vosa_usecase2.txt r1 manage 0.2 K 2011-03-17 - 14:40 UnknownUser VOSA. Use Case #2. Data
Unknown file formatEXT vosa_usecase3 r3 r2 r1 manage 0.2 K 2011-03-17 - 22:34 UnknownUser  
Texttxt vosa_usecase3.txt r1 manage 0.2 K 2011-03-17 - 22:24 UnknownUser  
Texttxt vosa_usecase3.txt.txt r1 manage 0.2 K 2011-03-17 - 22:21 UnknownUser VOSA. Use Case #3. Data
Unknown file formatEXT vosa_usecase3b r2 r1 manage 0.2 K 2011-03-17 - 22:56 UnknownUser  
Edit | Attach | Watch | Print version | History: r9 < r8 < r7 < r6 < r5 | Backlinks | Raw View | Raw edit | More topic actions
Topic revision: r9 - 2011-11-28 - EnriqueSolano
 
This site is powered by the TWiki collaboration platform Powered by Perl This site is powered by the TWiki collaboration platformCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback