IAU Symposium 241: Stellar Population Challenge

Organizer: Scott Trager, Kapteyn Astronomical Institute, University of Groningen

The purpose of the challenge is to test whether modern, "stock" (off-the-shelf, available freely to everyone) stellar population model ingredients --- stellar evolution models (isochrones and lifetimes) and stellar atmosphere models and observations (empirical and theoretical colour-Teff relations and bolometric corrections, synthetic and empirical stellar spectra, and fitting functions) --- can be combined to accurately reproduce low-resolution spectra of well-studied local objects.  In order to do this, the challenge is split into two sections that need to be done in series:

Test 1: Resolved Stellar Populations

Part 1: Matching the CMDs of a number of star (open and globular) clusters which have well-observed CMDs AND (low-resolution) spectra.  This first part tests the state of current isochrones and colour-Teff relations.  The key here is not to measure only an age for the objects, but to attempt to fit as best as possible the entire (observed) CMD within the capabilities of currently-available stellar evolution calculations.

Part 2: Matching the LFs of the objects (when available).  This tests the evolutionary lifetimes and the bolometric corrections of the stellar evolution models.

The deadline for this first test is 29 September.  The deliverables are the "best" tracks, isochrones, colour-Teff, and bolometric corrections for these 3 or 4 objects, and their ages as derived from these ingredients.  The definition of "best" is up to those participating in this part of the challenge, but matching the entire observed CMD is part of the definition.  However, the ground rules are that published or widely-available (i.e., on the Web) stellar evolution models should be used.  If non-standard models are used, they must be provided to the challenge organizer for use in Test 2 along with a coherent explanation of why it was necessary to use these models.

Results of Test 1:

Li et al.
Vazdekis et al.

Test 2: Unresolved Stellar Populations

This depends on the results of Test 1.  Stellar population modelers will use the "best" ingredients from Test1 and their own models and other ingredients (stellar atmospheres: synthetic spectra/fitting functions; IMF; etc.) --- but again, these ingredients must be published or widely-available (on the Web) or provided the challenge organizer--- to construct best-fitting spectra and/or absorption-line strengths of the objects from Test 1.  The deliverables are the ages and chemical compositions (at least metallicity and [alpha/Fe] or individual abundance ratios) for these objects.  The starting date for Test 2 is 9 October (to give the challenge organizer a week to post the results of Test 1) and the deadline is 1 December.

Results of Test 2:

Martins, Coelho & Cid Fernandes (PDF)
Vazdekis et al.


CMD reference
LF reference
Spectrum reference
Resolution (FWHM)
NGC 1805
Johnson et al. (2001); de Grijs & Liu, priv. comm. (ReadMe)
de Grijs et al. (2002a)
blue red
Santos et al. (2002)
12-16 Ang (blue); 14 Ang (red)
NGC 1868
Kerber & Santiago (2005); de Grijs & Liu, priv. comm. (ReadMe) here de Grijs et al. (2002b)
Santos et al. (2002)
11 Ang
Montgomery et al. (1993)

optical indexes
Schiavon et al. (2004)
2.7 Ang
NGC 6528
Feltzing & Johnson (2002)

Schiavon et al. (2005)
3.1 Ang
47 Tuc
Howell et al. (2000) (note: do not use U band photometry)

Schiavon et al. (2005) 3.1 Ang

Note (16 October 2006): NGC 6528 CMD has been updated with "cleaned" CMD from Feltzing & Johnson (2002).  Thanks to Sofia Feltzing for the data files.

Test 3: Extra Credit!

For those who do not want to/cannot participate in Tests 1 and 2 or want an additional challenge, we have prepared spectra of a number of objects with star formation histories (mostly!) known to the challenge organizer.  The deliverables for this test are the star formation histories, chemical compositions, and a description of the technique used to determine these. The deadline for Test 3 is 1 December.

Resolution (FWHM)
Spectrum 1 (ascii)
3 Ang
Spectrum 2 (ascii)
3 Ang
Spectrum 3 (ascii) 2.4 Ang
Spectrum 4 (ascii), error spectrum for Spectrum 4 (ascii)
170 km/s
Spectrum 5 (ascii)
2.7 Ang
Spectrum 6 (ascii) 3 Ang
Spectrum 7 (ascii) 2.4 Ang

Notes: (1) most but not all of these spectra have an intrinsic velocity dispersion; (2) flux scales are arbitrary; (3) wavelength scales are air wavelengths except for spectrum4 (although the ASCII version is now air wavelengths for consistency); (4) all wavelength scales are in log lambda/Angstrom except for spectrum5, which is in linear wavelength (Angstrom).

Results of Test 3:

Gomes & Cid Fernandes
Magris, Bruzual & Mateu

Challenge Results

Credit where credit is due:
Many thanks to Alexandre Vazdekis, Reynier Peletier, and the SOC of IAU Symp. 241; Maurizio Salaris, Ricardo Schiavon, Jim Rose, Lucimara Martins, Uta Frize, and the participants of the Fine-Tuning Stellar Population Models workshop; Richard de Grijs; and Sofia Feltzing.