Sub one per cent mass fractions of young stars in red massive galaxies


Early-type galaxies are considered to be the end products of massive galaxy formation1. Optical spectroscopic studies reveal that massive early-type galaxies formed the bulk of their stars over short timescales (\(\lesssim\)1 Gyr) and at high redshift (z\(\gtrsim\) 2), followed by passive evolution to the present2. However, their optical spectra are unable to constrain small episodes of recent star formation, since they are dominated by old stars. Fortunately, this problem can be tackled in the ultraviolet range. While recent studies that make use of ultraviolet absorption lines have suggested the presence of young stars in a few early-type galaxies3, the age and mass fractions of young stars and their dependence on galaxy mass are unknown. Here we report a detailed study of these young stellar populations, from high-quality stacked spectra of 28,663 galaxies from the BOSS survey4, analysing optical and ultraviolet absorption lines simultaneously. We find that residual star formation is ubiquitous in massive early-type galaxies, measuring average mass fractions of 0.5% in young stars in the last 2 Gyr of their evolution. This fraction shows a decreasing trend with galaxy stellar mass, consistent with a downsizing scenario5. We also find that synthetic galaxies from state-of-the-art cosmological numerical simulations6 substantially overproduce both intermediate and young stellar populations. Therefore, our results pose stringent constraints on numerical simulations of galaxy formation6,7.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Fig. 1
Fig. 2
Fig. 3

Data availability

The observational data used in this study are derived from the SDSS-III DR12 Survey Science Archive Server ( The stacked spectrum data analysed during the current study that support the plots within this paper are available as Source Data. The EAGLE simulation data are publicly available ( and the simulated spectra of the EAGLE galaxies analysed in this manuscript are available from the authors on reasonable request. The E-MILES SSP models are publicly available at the MILES website (

Code availability

To measure line-strength indices of the spectra we use our own code based on the publicly available indexf code from The emcee package used to fit the observed line-strength indices is available at Codes generated in this study are not publicly available since they can be reproduced following the Methods description but are available from the corresponding author on reasonable request.


  1. 1.

    Ferreras, I. & Silk, J. How young are early-type cluster galaxies? Quantifying the young stellar component in a rich cluster at z=0.41. Astrophys. J. 541, L37–L40 (2000).

    ADS  Google Scholar 

  2. 2.

    Renzini, A. Stellar population diagnostics of elliptical galaxy formation. Annu. Rev. Astron. Astrophys. 44, 141–192 (2006).

    ADS  Google Scholar 

  3. 3.

    Vazdekis, A., Koleva, M., Ricciardelli, E., Röck, B. & Falcón-Barroso, J. UV-extended E-MILES stellar population models: young components in massive early-type galaxies. Mon. Not. R. Astron. Soc. 463, 3409–3436 (2016).

    ADS  Google Scholar 

  4. 4.

    Dawson, K. S. et al. The Baryon Oscillation Spectroscopic Survey of SDSS-III. Astron. J. 145, 10–50 (2013).

    ADS  Google Scholar 

  5. 5.

    Cimatti, A., Daddi, E. & Renzini, A. Mass downsizing and “top-down” assembly of early-type galaxies. Astron. Astrophys. 453, L29–L33 (2006).

    ADS  Google Scholar 

  6. 6.

    Schaye, J. et al. The EAGLE project: simulating the evolution and assembly of galaxies and their environments. Mon. Not. R. Astron. Soc. 446, 521–554 (2015).

    ADS  Google Scholar 

  7. 7.

    Vogelsberger, M. et al. Introducing the Illustris Project: simulating the coevolution of dark and visible matter in the universe. Mon. Not. R. Astron. Soc. 444, 1518–1547 (2014).

    ADS  Google Scholar 

  8. 8.

    Kaviraj, S. et al. UV-optical colors as probes of early-type galaxy evolution. Astrophys. J. Suppl. Ser. 173, 619–642 (2007).

    ADS  Google Scholar 

  9. 9.

    Masters, K. L. et al. The morphology of galaxies in the Baryon Oscillation Spectroscopic Survey. Mon. Not. R. Astron. Soc. 418, 1055–1070 (2011).

    ADS  Google Scholar 

  10. 10.

    Bernardi, M., Sheth, R. K., Nichol, R. C., Schneider, D. P. & Brinkmann, J. Colors, magnitudes, and velocity dispersions in early-type galaxies: implications for galaxy ages and metallicities. Astron. J. 129, 61–72 (2005).

    ADS  Google Scholar 

  11. 11.

    La Barbera, F. et al. SPIDER VIII—constraints on the stellar initial mass function of early-type galaxies from a variety of spectral features. Mon. Not. R. Astron. Soc. 433, 3017–3047 (2013).

    ADS  Google Scholar 

  12. 12.

    Faber, S. M. & Jackson, R. E. Velocity dispersions and mass-to-light ratios for elliptical galaxies. Astrophys. J. 204, 668–683 (1976).

    ADS  Google Scholar 

  13. 13.

    Zahid, H. J., Geller, M. J., Fabricant, D. G. & Hwang, H. S. The scaling of stellar mass and central stellar velocity dispersion for quiescent galaxies at \(\rm{z}<0.7\). Astrophys. J. 832, 203 (2016).

    ADS  Google Scholar 

  14. 14.

    Le Cras, C., Maraston, C., Thomas, D. & Donald G., Y. Modelling the UV spectrum of SDSS-III/BOSS galaxies: hints towards the detection of the UV upturn at high-z. Mon. Not. R. Astron. Soc. 461, 766–793 (2016).

    ADS  Google Scholar 

  15. 15.

    Thomas, D., Maraston, C., Bender, R. & Mendes de Oliveira, C. The epochs of early-type galaxy formation as a function of environment. Astrophys. J. 621, 673–694 (2005).

    ADS  Google Scholar 

  16. 16.

    van der Wel, A. et al. The VLT LEGA-C spectroscopic survey: the physics of galaxies at a lookback time of 7 Gyr. Astrophys. J. Suppl. Ser. 223, 29 (2016).

    ADS  Google Scholar 

  17. 17.

    Wu, P. F. et al. Fast and slow paths to quiescence: ages and sizes of 400 quiescent galaxies from the LEGA-C survey. Astrophys. J. 868, 37 (2018).

    ADS  Google Scholar 

  18. 18.

    Cappellari, M. et al. The ATLAS3D project—I. A volume-limited sample of 260 nearby early-type galaxies: science goals and selection criteria. Mon. Not. R. Astron. Soc. 413, 813–836 (2011).

    ADS  Google Scholar 

  19. 19.

    McDermid, R. M. et al. The ATLAS3D project—XXX. Star formation histories and stellar population scaling relations of early-type galaxies. Mon. Not. R. Astron. Soc. 448, 3484–3513 (2015).

    ADS  Google Scholar 

  20. 20.

    Charlot, S. & Bruzual, A. G. Stellar population synthesis revisited. Astrophys. J. 367, 126–140 (1991).

    ADS  Google Scholar 

  21. 21.

    Conroy, C. Modeling the panchromatic spectral energy distributions of galaxies. Annu. Rev. Astron. Astrophys. 51, 393–455 (2013).

    ADS  Google Scholar 

  22. 22.

    Vidal Garcia, D., Maraston, C., Bender, R. & Mendes de Oliveira, C. The epochs of early-type galaxy formation as a function of environment. Astrophys. J. 621, 673–694 (2017).

    Google Scholar 

  23. 23.

    Gabor, J. M., Dave, R., Finlator, K. & Oppenheimer, B. D. How is star formation quenched in massive galaxies? Mon. Not. R. Astron. Soc. 407, 749–771 (2010).

    ADS  Google Scholar 

  24. 24.

    Springel, V., Di Matteo, T. & Hernquist, L. Black holes in galaxy mergers: the formation of red elliptical galaxies. Astrophys. J. 620, L79–L82 (2005).

    ADS  Google Scholar 

  25. 25.

    Schawinski, K. et al. Suppression of star formation in early-type galaxies by feedback from supermassive black holes. Nature 442, 888–891 (2006).

    ADS  Google Scholar 

  26. 26.

    Dubois, Y., Gavazzi, R., Peirani, S. & Silk, J. AGN-driven quenching of star formation: morphological and dynamical implications for early-type galaxies. Mon. Not. R. Astron. Soc. 433, 3297–3313 (2013).

    ADS  Google Scholar 

  27. 27.

    Welch, G. A., Sage, L. J. & Young, L. M. The cool interstellar medium in elliptical galaxies. II. Gas content in the volume-limited sample and results from the combined elliptical and lenticular surveys. Astrophys. J. 725, 100–114 (2010).

    ADS  Google Scholar 

  28. 28.

    Young, L. M. et al. The ATLAS3D project—XXVII. Cold gas and the colours and ages of early-type galaxies. Mon. Not. R. Astron. Soc. 444, 3408–3426 (2014).

    ADS  Google Scholar 

  29. 29.

    Maraston, C. et al. Stellar masses of SDSS-III/BOSS galaxies at z 0.5 and constraints to galaxy formation models. Mon. Not. R. Astron. Soc. 435, 2764–2792 (2013).

    ADS  Google Scholar 

  30. 30.

    Vazdekis, A., Peletier, R. F., Beckman, J. E. & Casuso, E. A new chemo-evolutionary population synthesis model for early-type galaxies. II. Observations and results. Astrophys. J. 111, 203–232 (1997).

    ADS  Google Scholar 

  31. 31.

    La Barbera, F. et al. SPIDER—I. Sample and galaxy parameters in the grizYJHK wavebands. Mon. Not. R. Astron. Soc. 408, 1313–1334 (2010).

    ADS  Google Scholar 

  32. 32.

    Cappellari, M. & Ensellem, E. Parametric recovery of line-of-sight velocity distributions from absorption-line spectra of galaxies via penalized likelihood. Publ. Astron. Soc. Pac. 116, 138–147 (2004).

    ADS  Google Scholar 

  33. 33.

    Thomas, D. et al. Stellar velocity dispersions and emission line properties of SDSS-III/BOSS galaxies. Mon. Not. R. Astron. Soc. 431, 1383–1397 (2013).

    ADS  Google Scholar 

  34. 34.

    Morton, D. C. Atomic data for resonance absorption lines. I. Wavelengths longward of the Lyman limit. Astrophys. J. Suppl. Ser. 77, 119–202 (1991).

    ADS  Google Scholar 

  35. 35.

    Cardelli, J., Clayton, G. C. & Mathis, J. S. The relationship between infrared, optical, and ultraviolet extinction. Astrophys. J. 345, 245–256 (1989).

    ADS  Google Scholar 

  36. 36.

    Stoughton et al. Sloan Digital Sky Survey: early data release. Astron. J. 123, 485–548 (2002).

    ADS  Google Scholar 

  37. 37.

    Gorgas, J., Pedraz, S., Guzmán, R., Cardiel, N. & González, J. J. Line-strength indices in bright spheroidal galaxies: evidence for a stellar population dichotomy between spheroidal and elliptical galaxies. Astrophys. J. 481, L19–L22 (1997).

    ADS  Google Scholar 

  38. 38.

    Sánchez-Blázquez, P. et al. Stellar populations of early-type galaxies in different environments. I. Line-strength indices. Relations of line-strengths with σ. Astron. Astrophys. 457, 787–808 (2006).

    ADS  Google Scholar 

  39. 39.

    Vazdekis, A., Trujillo, I. & Yamada, Y. A correlation between light profile and [Mg/Fe] abundance ratio in early-type galaxies. Astrophys. J. 601, L33–L36 (2004).

    ADS  Google Scholar 

  40. 40.

    Asa’d, R. et al. Young LMC clusters: the role of red supergiants and multiple stellar populations in their integrated light and CMDs. Mon. Not. R. Astron. Soc. 471, 3599–3614 (2017).

    ADS  Google Scholar 

  41. 41.

    Girardi, L., Bressan, A., Bertelli, G. & Chiosi, C. Evolutionary tracks and isochrones for low- and intermediate-mass stars: From 0.15 to 7 M , and from Z=0.0004 to 0.03. Astrophys. J. Suppl. Ser. 141, 371–383 (2000).

    ADS  Google Scholar 

  42. 42.

    Vazdekis, A., Casuso, E., Peletier, R. F. & Beckman, J. E. A new chemo-evolutionary population synthesis model for early-type galaxies. I. Theoretical basis. Astrophys. J. 106, 307 (1996).

    ADS  Google Scholar 

  43. 43.

    Press, W. H., Teukolsky, S. A., Vetterling, W. T. & Flannery, B. P. Numerical Recipes 3rd edn (Cambridge Univ. Press, 2007).

  44. 44.

    Planck Collaboration et al. Planck 2013 results. XVI. Cosmological parameters. Astron. Astrophys. 571, A16 (2014).

  45. 45.

    Patil, M. K., Pandey, S. K., Sahu, D. K. & Kembhavi, A. Properties of dust in early-type galaxies. Astron. Astrophys. 461, 103–113 (2007).

    ADS  Google Scholar 

  46. 46.

    de la Rosa, I. G., La Barbera, F., Ferreras, I. & de Carvalho, R. R. The link between the star formation history and [α/Fe]. Mon. Not. R. Astron. Soc. 418, L74–L78 (2011).

    ADS  Google Scholar 

  47. 47.

    Foreman-Mackey, D., Hogg, D. W., Lang, D. & Goodman, J. emcee: the MCMC hammer. Publ. Astron. Soc. Pac. 125, 306 (2013).

    ADS  Google Scholar 

  48. 48.

    Goodman, J. & Weare, J. Ensemble samplers with affine invariance. Commun. Appl. Math. Comput. Sci. 5, 65–80 (2010).

    MathSciNet  MATH  Google Scholar 

  49. 49.

    Serven, J., Worthey, G., Toloba, E. & Sánchez-Blázquez, P. NH and Mg index trends in elliptical galaxies. Astron. J. 141, 184 (2011).

    ADS  Google Scholar 

  50. 50.

    Buzzoni, A. Evolutionary population synthesis in stellar systems. I—A global approach. Astrophys.J. Suppl. Ser. 71, 817–869 (1989).

    ADS  Google Scholar 

  51. 51.

    Schwarz, G. E. Estimating the dimension of a model. Ann. Stat. 6, 461–464 (1978).

    MathSciNet  MATH  Google Scholar 

  52. 52.

    Moster, B. P. et al. Constraints on the relationship between stellar mass and halo mass at low and high redshift. Astrophys. J. 710, 903–923 (2010).

    ADS  Google Scholar 

  53. 53.

    Guo, Q., White, S., Li, C. & Boylan-Kolchin, M. How do galaxies populate dark matter haloes? Mon. Not. R. Astron. Soc. 404, 1111–1120 (2010).

    ADS  Google Scholar 

  54. 54.

    Behroozi, P. S., Conroy, C. & Wechsler, R. H. A comprehensive analysis of uncertainties affecting the stellar mass–halo mass relation for \(0<{\rm{z}}<4\). Astrophys. J. 717, 379–403 (2010).

    ADS  Google Scholar 

  55. 55.

    McAlpine, S. et al. The EAGLE simulations of galaxy formation: public release of halo and galaxy catalogues. Astron. Comput. 15, 72–89 (2016).

    ADS  Google Scholar 

  56. 56.

    Schaye, J. & Dalla Vecchia, C. On the relation between the Schmidt and Kennicutt–Schmidt star formation laws and its implications for numerical simulations. Mon. Not. R. Astron. Soc. 383, 1210–1222 (2008).

    ADS  Google Scholar 

  57. 57.

    Wiersma, R. P. C., Schaye, J. & Smith, B. D. The effect of photoionization on the cooling rates of enriched, astrophysical plasmas. Mon. Not. R. Astron. Soc. 393, 99–107 (2009).

    ADS  Google Scholar 

  58. 58.

    Wiersma, R. P. C., Schaye, J., Theuns, T., Dalla Vecchia, C. & Tornatore, L. Chemical enrichment in cosmological, smoothed particle hydrodynamics simulations. Mon. Not. R. Astron. Soc. 399, 574–600 (2009).

    ADS  Google Scholar 

  59. 59.

    Dalla Vecchia, C. & Schaye, J. Simulating galactic outflows with thermal supernova feedback. Mon. Not. R. Astron. Soc. 426, 140–158 (2012).

    ADS  Google Scholar 

  60. 60.

    Rosas-Guevara, Y. M. et al. The impact of angular momentum on black hole accretion rates in simulations of galaxy formation. Mon. Not. R. Astron. Soc. 454, 1038–1057 (2015).

    ADS  Google Scholar 

  61. 61.

    Crain, R. A. et al. The EAGLE simulations of galaxy formation: calibration of subgrid physics and model variations. Mon. Not. R. Astron. Soc. 450, 1937–1961 (2015).

    ADS  Google Scholar 

  62. 62.

    Schaller, M. et al. The EAGLE simulations of galaxy formation: the importance of the hydrodynamics scheme. Mon. Not. R. Astron. Soc. 454, 2277–2291 (2015).

    ADS  Google Scholar 

  63. 63.

    Trayford, J. W. et al. Colours and luminosities of z = 0.1 galaxies in the EAGLE simulation. Mon. Not. R. Astron. Soc. 452, 2879–2896 (2015).

    ADS  Google Scholar 

Download references


Funding for SDSS-III has been provided by the Alfred P. Sloan Foundation, the Participating Institutions, the National Science Foundation and the US Department of Energy Office of Science. The SDSS-III website is SDSS-III is managed by the Astrophysical Research Consortium for the Participating Institutions of the SDSS-III Collaboration including the University of Arizona, the Brazilian Participation Group, Brookhaven National Laboratory, Carnegie Mellon University, University of Florida, the French Participation Group, the German Participation Group, Harvard University, the Instituto de Astrofísica de Canarias, the Michigan State/Notre Dame/JINA Participation Group, Johns Hopkins University, Lawrence Berkeley National Laboratory, Max Planck Institute for Astrophysics, Max Planck Institute for Extraterrestrial Physics, New Mexico State University, New York University, Ohio State University, Pennsylvania State University, University of Portsmouth, Princeton University, the Spanish Participation Group, University of Tokyo, University of Utah, Vanderbilt University, University of Virginia, University of Washington and Yale University. We thank G. Bruzual and S. Charlot for kindly providing us with their new version of stellar population models extended to the ultraviolet spectral range to test how results change when using other SSP models. Parts of this research have been funded by the Spanish Ministry of Science, Innovation and Universities (MCIU) through research project SEV-2015-0548-16-4 and predoctoral contract BES-2016-078409. N.S.R., M.A.B., A.V. and I.F. acknowledge support from grant AYA2016-77237-C3-1-P from the MCIU. F.L.B. acknowledges financial support from the European Union’s Horizon 2020 research and innovation programme under Marie Sklodowska-Curie grant agreement 721463 to the SUNDIAL ITN network. M.A.B. and A.N. acknowledge the Severo Ochoa excellence scheme SEV-2015-0548. C.D.V. acknowledges financial support from MCIU through grants AYA2014-58308 and RYC-2015-1807.

Author information




N.S.R. led the analysis and wrote the text. A.V., M.A.B. and F.L.B. developed the idea and supervised the work. I.F. prepared the BOSS data for analysis. F.L.B and N.S.R. wrote code. N.S.R., A.V., M.A.B., F.L.B. and I.F. interpreted and analysed the results. A.N. and C.D.V. prepared the simulation data. All authors discussed the results and edited the manuscript.

Corresponding author

Correspondence to Núria Salvador-Rusiñol.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Peer review information Nature Astronomy thanks Camilla Pacifici and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 Data selection criteria.

Selection of our BOSS galaxy sample as a function of velocity dispersion and redshift. The plot illustrates the number density in logarithmic scale of available BOSS spectra with SNR > 7 in the SDSS-r band and (g-i) > 2.35. We select BOSS galaxies with redshift \(0.35{\le }{{\rm{z}}}\le0.6\), and velocity dispersion between 220 and 340 km s\({}^{-1}\). Lowest density regions show individual galaxies with dark blue.

Extended Data Fig. 2 BOSS stacked spectra.

Panel a) shows our stacked spectra from the BOSS survey, colour-coded with respect to the stellar velocity dispersion range labelled on the top left of each spectrum. The spectra have been shifted vertically for ease of visualization. Panel b) represents the signal-to-noise ratio, following the same colour scheme.

Extended Data Fig. 3 Probability distribution functions from MCMC.

Probability distribution functions of an intermediate velocity dispersion stack (250–260 km/s) corresponding to the model parameters obtained with the 2SSPs approach. From top to bottom: age and metallicity for the old component, and age and mass fraction of the young component. The panels show the marginalized PDF over each single parameter over the other parameters. Contours show the two-dimensional \(1\sigma\), \(2\sigma\) and \(3\sigma\) confidence levels between all the parameters. The last panel of each row shows the median (blue solid line) and the 16 and 84 percentiles (dashed lines) for each PDF. Note the burst age–burst strength degeneracy found between the age and the fraction of the young component, for which a smaller burst can be produced more recently.

Extended Data Fig. 4 NUV and optical line-strength indices.

The panels show the behaviour of the NUV and optical SSP model indices as a function of age, for two different metallicities ([M/H]=0.0, solid lines; [M/H]=+0.22, dashed lines). The black dots correspond to the observed measurements of BOSS stacked spectra and their error bars show the 1𝜎 uncertainty on the measurements. Red stars indicate the best fitting 2SSPs model predictions. Blue triangles show the index values predicted by the best fitting 1SSP+cSFR model. Note that the discrepancy between SSP model predictions and data increases towards bluer indices. The indices shown exclude those in Figure 1.

Extended Data Fig. 5 2SSPs best index fit spectrum.

Panel a) shows the stacked spectrum with the highest velocity dispersion range (thick black line). Overplotted is the 2SSPs model (thin black) that best fits the 14 indices shown in Figure 1 and in Extended Data 4. The main purpose of this figure is to illustrate the relative effect of the young component as a function of wavelength but note it does not represent a full-spectrum fit. The 2SSPs model can be split into the young (65 Myr) component (blue dashed) which contributes a 0.09% mass fraction, and the old (9.38 Gyr) dominant population (red dotted). Panel b) shows the ratio between the 2SSPs best fit model with the stacked spectrum (black) and with a pure old SSP model of 9.38 Gyr (orange). The inset subplot shows the flux ratio of the dashed box in more detail.

Supplementary information

Supplementary Information

Supplementary Tables 1 and 2.

Source data

Source Data for Extended Data Fig. 2

BOSS stacked spectra data sorted by the velocity dispersion of each bin, smoothed at 340 km s−1.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Salvador-Rusiñol, N., Vazdekis, A., La Barbera, F. et al. Sub one per cent mass fractions of young stars in red massive galaxies. Nat Astron 4, 252–259 (2020).

Download citation

Further reading


Quick links

Sign up for the Nature Briefing newsletter for a daily update on COVID-19 science.
Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing