Better age estimations using UV-optical colours: breaking the age-metallicity degeneracy

Department of Physics, University of Oxford, Oxford, England, United Kingdom
Monthly Notices of the Royal Astronomical Society (Impact Factor: 5.11). 01/2006; 381(1). DOI: 10.1111/j.1745-3933.2007.00370.x
Source: arXiv


We demonstrate that the combination of GALEX UV photometry in the FUV (~1530 angstroms) and NUV (~2310 angstroms) passbands with optical photometry in the standard U,B,V,R,I filters can efficiently break the age-metallicity degeneracy. We estimate well-constrained ages, metallicities and their associated errors for 42 GCs in M31, and show that the full set of FUV,NUV,U,B,V,R,I photometry produces age estimates that are ~90 percent more constrained and metallicity estimates that are ~60 percent more constrained than those produced by using optical filters alone. The quality of the age constraints is comparable or marginally better than those achieved using a large number of spectrscopic indices. Comment: Published in MNRAS (2007), 381, L74 (doi: 10.1111/j.1745-3933.2007.00370.x)

Download full-text


Available from: Suk-Jin Yoon, Apr 29, 2013
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Using a widely used stellar-population synthesis model, we study the possibility of using pairs of AB system colours to break the well-known stellar age–metallicity degeneracy and to give constraints on two luminosity-weighted stellar-population parameters (age and metallicity). We present the relative age and metallicity sensitivities of the AB system colours that relate to the u, B, g, V, r, R, i, I, z, J, H and K bands, and we quantify the ability of various colour pairs to break the age–metallicity degeneracy. Our results suggest that a few pairs of colours can be used to constrain the above two stellar-population parameters. This will be very useful for exploring the stellar populations of distant galaxies. In detail, colour pairs [(r–K), (u–R)] and [(r–K), (u–r)] are shown to be the best pairs for estimating the luminosity-weighted stellar ages and metallicities of galaxies. They can constrain two stellar-population parameters on average with age uncertainties less than 3.89 Gyr and metallicity uncertainties less than 0.34 dex for typical colour uncertainties. The typical age uncertainties for young populations (age < 4.6 Gyr) and metal-rich populations (Z≥ 0.001) are small (about 2.26 Gyr) while those for old populations (age ≥ 4.6 Gyr) and metal-poor populations (Z < 0.001) are much larger (about 6.88 Gyr). However, the metallicity uncertainties for metal-poor populations (about 0.0024) are much smaller than for other populations (about 0.015). Some other colour pairs can also possibly be used for constraining the two parameters. On the whole, the estimation of stellar-population parameters is likely to be reliable only for early-type galaxies with small colour errors and globular clusters, because such objects contain less dust. In fact, no galaxy is totally dust-free and early-type galaxies are also likely have some dust [e.g. E(B–V) ∼ 0.05], which can change the stellar ages by about 2.5 Gyr and metallicities (Z) by about 0.015. When we compare the photometric estimates with previous spectroscopic estimates, we find some differences, especially when comparing the stellar ages determined by two methods. The differences mainly result from the young populations of galaxies. Therefore, it is difficult to obtain the absolute values of stellar ages and metallicities, but the results are useful for obtaining some relative values. In addition, our results suggest that colours relating to both UBVRIJHK and ugriz magnitudes are much better than either UBVRIJHK or ugriz colours for breaking the well-known degeneracy. The results also show that the stellar ages and metallicities of galaxies observed by the Sloan Digital Sky Survey and the Two-Micron All-Sky Survey can be estimated via photometry data.
    Monthly Notices of the Royal Astronomical Society 04/2007; 385(3). DOI:10.1111/j.1365-2966.2008.12661.x · 5.11 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We demonstrate that the large scatter in the ultra-violet (UV) colours of intermediate-mass early-type galaxies in the local Universe and the inferred low-level recent star formation in these objects can be reproduced by minor mergers in the standard LCDM cosmology. Numerical simulations of mergers with mass ratios less than or equal to 1:4, with reasonable assumptions for the ages, metallicities and dust properties of the merger progenitors, produce good agreement to the observed UV colours of the early-type population, if the infalling satellites are assumed to have (cold) gas fractions greater than 20%. Early-types that satisfy (NUV-r) < 3.8 are likely to have experienced mergers with mass ratios between 1:4 and 1:6 within the last ~1.5 Gyrs, while those that satisfy 3.8<(NUV-r)<5.5 are consistent with either recent mergers with mass ratios < 1:6 or mergers with higher mass ratios that occurred more than ~1.5 Gyrs in the past. We demonstrate that the early-type colour-magnitude relations and colour distributions in both the UV and optical spectral ranges are consistent with the expected frequency of minor merging activity in the standard LCDM cosmology at low redshift. We present a strong plausibility argument for minor mergers to be the principal mechanism behind the large UV scatter and associated low-level recent star formation observed in early-type galaxies in the nearby Universe. Comment: MNRAS in press
    Monthly Notices of the Royal Astronomical Society 11/2007; 394(4). DOI:10.1111/j.1365-2966.2009.14403.x · 5.11 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We determine the age of 1,104 early-type galaxies in eight rich clusters ($z = 0.0046$ to $0.175$) using a new continuum color technique. We find that galaxies in clusters divide into two populations, an old population with a mean age similar to the age of the Universe (12 Gyrs) and a younger population with a mean age of 9 Gyrs. The older population follows the expected relations for mass and metallicity that imply a classic monolithic collapse origin. Although total galaxy metallicity is correlated with galaxy mass, it is uncorrelated with age. It is impossible, with the current data, to distinguish between a later epoch of star formation, longer duration of star formation or late bursts of star formation to explain the difference between the old and young populations. However, the global properties of this younger population are correlated with cluster environmental factors, which implies secondary processes, post-formation epoch, operate on the internal stellar population of a significant fraction of cluster galaxies. In addition, the mean age of the oldest galaxies in a cluster are correlated with cluster velocity dispersion implying that galaxy formation in massive clusters begins at earlier epochs than less massive clusters. Comment: 35 pages, 10 figures, accepted by ApJ
    The Astrophysical Journal 01/2008; 677(2). DOI:10.1086/533513 · 5.99 Impact Factor
Show more