No Arabic abstract
We have obtained the mass-metallicity (M-Z) relation at different lookback times for the same set of galaxies from the Sloan Digital Sky Survey, using the stellar metallicities estimated with our spectral synthesis code STARLIGHT. We have found that this relation steepens and spans a wider range in both mass and metallicity at higher redshifts. We have modeled the time evolution of stellar metallicity with a closed-box chemical evolution model, for galaxies of different types and masses. Our results suggest that the M-Z relation for galaxies with present-day stellar masses down to 10^10 M_sun is mainly driven by the history of star formation history and not by inflows or outflows.
We use fossil record techniques on the CALIFA sample to study how galaxies in the local universe have evolved in terms of their chemical content. We show how the metallicity and the mass-metallicity relation (MZR) evolve through time for the galaxies in our sample and how this evolution varies when we divide them based on their mass, morphology and star-forming status. We also check the impact of measuring the metallicity at the centre or the outskirts. We find the expected results that the most massive galaxies got enriched faster, with the MZR getting steeper at higher redshifts. However, once we separate the galaxies into morphology bins this behaviour is not as clear, which suggests that morphology is a primary factor to determine how fast a galaxy gets enriched, with mass determining the amount of enrichment. We also find that star-forming galaxies appear to be converging in their chemical evolution, that is, the metallicity of star-forming galaxies of different mass is very similar at recent times compared to several Gyr ago.
Active galactic nuclei (AGNs) are characterized by a clear correlation between luminosity and metallicity (L_AGN-Z_AGN relation). The origin of this correlation is not clear. It may result from a relation between the black hole mass (M_BH) and metallicity, or from a relation between the accretion rate (L/L_Edd) and metallicity. To investigate the origin of the L_AGN-Z_AGN relation, we use optical spectra of 2383 quasars at 2.3 < z < 3.0 from the Sloan Digital Sky Survey. By using this data set, we have constructed composite spectra of 33 subsamples in intervals of both M_BH and L/L_Edd. From these composite spectra we measure emission-line flux ratios that are sensitive to the metallicity of the broad line region (BLR); specifically, NV/CIV, NV/HeII, (SiIV+OIV])/CIV, and AlIII/CIV. We find that there is a significant correlation between M_BH and Z_BLR as inferred from all four metallicity-sensitive emission-line flux ratios. This result strongly suggests that the observed L_AGN-Z_AGN relation is mostly a consequence of the M_BH-Z_AGN relation. The relation between M_BH and Z_BLR is likely a consequence of both the M_BH-M_bul relation and of the mass-metallicity relation in the host galaxy. We also find that L/L_Edd correlates with the emission line flux ratios involving NV (more specifically, NV/CIV and NV/HeII), while it does not correlate with the other two metallicity sensitive emission line flux ratios, i.e., (SiIV+OIV])/CIV and AlIII/CIV. These correlations indicate that the emission-line flux ratios involving NV depend on both metallicity and relative abundance of nitrogen. We suggest that the relation between L/L_Edd and those line ratios involving nitrogen, is caused by a delay of the black hole accretion rate relative to the onset of nuclear star formation of about 10^8 years, which is the timescale required for the nitrogen enrichment.
During the last three decades, many papers have reported the existence of a luminosity-metallicity or mass-metallicity (M-Z) relation for all kinds of galaxies: The more massive galaxies are also the ones with more metal-rich interstellar medium. We have obtained the mass-metallicity relation at different lookback times for the same set of galaxies from the Sloan Digital Sky Survey (SDSS), using the stellar metallicities estimated with our spectral synthesis code STARLIGHT. Using stellar metallicities has several advantages: We are free of the biases that affect the calibration of nebular metallicities; we can include in our study objects for which the nebular metallicity cannot be measured, such as AGN hosts and passive galaxies; we can probe metallicities at different epochs of a galaxy evolution. We have found that the M-Z relation steepens and spans a wider range in both mass and metallicity at higher redshifts for SDSS galaxies. We also have modeled the time evolution of stellar metallicity with a closed-box chemical evolution model, for galaxies of different types and masses. Our results suggest that the M-Z relation for galaxies with present-day stellar masses down to 10^10 solar masses is mainly driven by the star formation history and not by inflows or outflows.
Dwarf galaxies generally follow a mass-metallicity (MZ) relation, where more massive objects retain a larger fraction of heavy elements. Young tidal dwarf galaxies (TDGs), born in the tidal tails produced by interacting gas-rich galaxies, have been thought to not follow the MZ relation, because they inherit the metallicity of the more massive parent galaxies. We present chemical evolution models to investigate if TDGs that formed at very high redshifts, where the metallicity of their parent galaxy was very low, can produce the observed MZ relation. Assuming that galaxy interactions were more frequent in the denser high-redshift universe, TDGs could constitute an important contribution to the dwarf galaxy population. The survey of chemical evolution models of TDGs presented here captures for the first time an initial mass function (IMF) of stars that is dependent on both the star formation rate and the gas metallicity via the integrated galactic IMF (IGIMF) theory. As TDGs form in the tidal debris of interacting galaxies, the pre-enrichment of the gas, an underlying pre-existing stellar population, infall, and mass dependent outflows are considered. The models of young TDGs that are created in strongly pre-enriched tidal arms with a pre-existing stellar population can explain the measured abundance ratios of observed TDGs. The same chemical evolution models for TDGs, that form out of gas with initially very low metallicity, naturally build up the observed MZ relation. The modelled chemical composition of ancient TDGs is therefore consistent with the observed MZ relation of satellite galaxies.
We study the shape of the gas-phase mass-metallicity relation (MZR) of a combined sample of present-day dwarf and high-mass star-forming galaxies using IZI, a Bayesian formalism for measuring chemical abundances presented in Blanc et al. 2015. We observe a characteristic stellar mass scale at $M_* simeq 10^{9.5}$M$_{odot}$, above which the ISM undergoes a sharp increase in its level of chemical enrichment. In the $10^{6}-10^{9.5}$M$_{odot}$ range the MZR follows a shallow power-law ($Zpropto M^{alpha}_*$) with slope $alpha=0.14pm0.08$. At approaching $M_* simeq 10^{9.5}$M$_{odot}$ the MZR steepens significantly, showing a slope of $alpha=0.37pm0.08$ in the $10^{9.5}-10^{10.5}$M$_{odot}$ range, and a flattening towards a constant metallicity at higher stellar masses. This behavior is qualitatively different from results in the literature that show a single power-law MZR towards the low mass end. We thoroughly explore systematic uncertainties in our measurement, and show that the shape of the MZR is not induced by sample selection, aperture effects, a changing N/O abundance, the adopted methodology used to construct the MZR, secondary dependencies on star formation activity, nor diffuse ionized gas (DIG) contamination, but rather on differences in the method used to measure abundances. High resolution hydrodynamical simulations can qualitatively reproduce our result, and suggest a transition in the ability of galaxies to retain their metals for stellar masses above this threshold. The MZR characteristic mass scale also coincides with a transition in the scale height and clumpiness of cold gas disks, and a typical gas fraction below which the efficiency of star formation feedback for driving outflows is expected to decrease sharply.