The Star Formation Histories of galaxies: A tour through the STARLIGHT-SDSS database

The Star Formation Histories of galaxies: A tour through the STARLIGHT-SDSS database

R. Cid Fernandes,1 W. Schoenell,1 J. M. Gomes,1 N V. Asari,1 M. Schlickmann,1 A. Mateus,2 G. Stasińska,3 L. Sodré,4 and J. P. Torres-Papaqui5 (the SEAGal collaboration)
Abstract

Retrieving the Star Formation History (SFH) of a galaxy out of its integrated spectrum is the central goal of stellar population synthesis. Recent advances in evolutionary synthesis models have given new breath to this old field of research. Modern spectral synthesis techniques incorporating these advances now allow the fitting of galaxy spectra on an Å-by-Å basis. These detailed fits are useful for a number of studies, like emission line, stellar kinematics, and specially galaxy evolution. Applications of this semi-empirical approach to mega data sets are teaching us a lot about the lives of galaxies. The STARLIGHT spectral synthesis code is one of the tools which allows one to harness this favorable combination of plentifulness of data and models. To illustrate this, we show how SFHs vary across classical emission line diagnostic diagrams. Systematic trends are present along both the star-forming and active-galaxy sequences. We also briefly describe experiments with new versions of evolutionary synthesis models. Last but not least, we announce the public availability of both STARLIGHT and a database of detailed spectral fits and related products for over half a million galaxies from the SDSS. This facility allows more physically inspired explorations of the parameter space than is possible in terms of raw observed properties, offering new ways to navigate through the realm of galaxies.

The Star Formation Histories of galaxies: A tour through the STARLIGHT-SDSS database

R. Cid Fernandes,1 W. Schoenell,1 J. M. Gomes,1 N V. Asari,1 M. Schlickmann,1 A. Mateus,2 G. Stasińska,3 L. Sodré,4 and J. P. Torres-Papaqui5 (the SEAGal collaboration)

11footnotetext: Departamento de Física - CFM, Universidade Federal de Santa Catarina, Florianópolis, SC, Brazil.22footnotetext: Laboratoire d’Astrophysique de Marseille, CNRS UMR6110, Traverse du Siphon, 13012 Marseille, France33footnotetext: LUTH, Observatoire de Paris, CNRS, Université Paris Diderot; Place Jules Janssen 92190 Meudon, France44footnotetext: Instituto de Astronomia, Geofísica e Ciências Atmosféricas, Universidade de São Paulo, Brazil55footnotetext: Instituto Nacional de Astrofísica, Óptica y Electrónica, Puebla, México

1 Introduction

How do galaxies assemble their stars over time, i.e, how do they evolve? This broad question lies at the heart of a large number of astrophysical frontier-problems, from the internal physics of galaxies to cosmological issues. Theoreticians try to answer this question plugging in as much physics as they can in their models (see Abadi’s talk), while many observers tackle it using the expanding Universe as a time-machine, examining how galaxy properties change with redshift (e.g., Aretxaga’s talk). This contribution deals with what can be learnt from a third and independent method: Uncovering the fossil record of evolution from its imprints on galaxy spectra. This semi-empirical approach has become highly attractive in the past few years, given the avalanche of data from cosmologically shallow surveys, and the enormous progress in our ability to fit galaxy spectra on a pixel-by-pixel basis using state-of-the-art evolutionary synthesis models as those reviewed in Bruzual’s and Coelho’s contributions.

Taking full advantage of this favorable combination of abundance of data and models requires spectral synthesis tools to extract information about age (), metallicity () and the detailed star formation history (SFH) encoded in observed spectra. There are now several such tools around, differing in both technical and astrophysical aspects (eg, Mateu’s talk). Some account for extinction and/or kinematics, others don’t. Some impose simple - relations or use a fixed , while others treat and independently. Some model indices, others the full spectrum. Some fit the SFH in a non-parametric fashion, while others compare the data to a library of precomputed models. Some prefer to compress the input data, others the output parameters. The list goes on and on…We skip the impossible task of covering all this ground by referring the reader to two reviews by our group (astroph/071899 and 071902), dedicated to basic aspects of spectral synthesis and the recent literature in the field.

These few pages focus on results obtained with our STARLIGHT synthesis code. Rather than self-advertisement, this is done with the specific purpose of illustrating the sort of science doable with this code, specially when applied to mega data sets. The motivation for this is that both STARLIGHT and products of its application to all SDSS galaxies are now available for public use. In www.starlight.ufsc.br the reader will find the code itself, Å-by-Å fits for 573141 galaxies from the SDSS DR5 and a long and diversified list of derived properties (from stellar masses to emission line fluxes). This database is about to double with 694135 DR6 spectra fitted with new evolutionary synthesis ingredients.

We start with a quick introduction to STARLIGHT, its deliverables and our VO-like database (§2), including a first ever comparison of results obtained with DR5 data modeled with the standard Bruzual & Charlot (2003) models with DR6 data fitted with newer models. As an invitation to our database, we illustrate the power of our detailed spectral synthesis approach by showing how SFHs vary as a function of location on classical emission line diagnostic diagrams (§LABEL:anon:sec:SFH_on_BPT).

2 The STARLIGHT/SEAGal Project

2.1 The code

STARLIGHT (Cid Fernandes et al. 2005) combines spectra from a user-defined base of individual populations in search of linear combinations which match an input observed spectrum. The fitted coefficients define an -dimensional population vector (light fractions at a reference ). For SFH studies it is useful to use a base of instantaneous bursts of different ’s and ’s, but anything else can be used. Kinematics is allowed for, as is reddening (according to any law). Papers by our Semi Empirical Analysis of Galaxies (SEAGal) collaboration discuss the code and the results of its application to 0.5M galaxies from the SDSS. For an in depth description of the code, its pros and cons and possible uses a detailed 45-pages long manual is available.

STARLIGHT itself outputs:

  • The full synthetic spectrum .

  • The light-fraction population vector .

  • A mass-fraction population vector (only meaningful for properly defined bases).

  • Stellar velocity dispersion () and shift ().

  • Stellar extinction ().

This is already enough for those interested in, say, stellar kinematics, or if all you need is a decent stellar template to subtract from your data to aid emission line work. In neither case you would care about nor . For galaxy evolution work, on the other hand, it is precisely the population vector which matters, since this is where the SFH information is. It can be handled in numerous ways to produce things like time dependent star formation rates, SFR(t).

2.2 The database: www.starlight.ufsc.br

Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
""
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
   
Add comment
Cancel
Loading ...
155885
This is a comment super asjknd jkasnjk adsnkj
Upvote
Downvote
""
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters
Submit
Cancel

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test
Test description