MASSIV: Mass Assembly Survey with SINFONI in VVDS

Abstract

Aims. The estimate of radial abundance gradients in high-redshift galaxies allows to constrain their star formation history and their interplay with the surrounding intergalactic medium. Methods. We present VLT/SINFONI integral-field spectroscopy of a first sample of 50 galaxies at z ∼ 1.2 in the MASSIV survey. Using the N2 ratio between the [N ii]6584 and Hα rest-frame optical emission lines as a proxy for oxygen abundance in the interstellar medium, we measured the metallicity of the sample galaxies. We developed a tool to extract spectra in annular regions, leading to a spatially resolved estimate of the oxygen abundance in each galaxy. We were able to derive a metallicity gradient for 26 galaxies in our sample and discovered a significant fraction of galaxies with a "positive" gradient. Using a simple chemical evolution model, we derived infall rates of pristine gas onto the disks. Results. Seven galaxies display a positive gradient at a high confidence level. Four out of these are interacting, and one is a chain galaxy. We suggest that interactions might be responsible for shallowing and even inverting the abundance gradient. We also identify two interesting correlations in our sample: a) galaxies with higher gas velocity dispersion have shallower/positive gradients; and b) metal-poor galaxies tend to show a positive gradient, whereas metal-rich ones tend to show a negative one. This last observation can be explained by the infall of metal-poor gas into the center of the disks. We address the question of the origin of this infall under the influence of gas flows triggered by interactions and/or cold gas accretion. All the data published in this paper are publicly available at the time of publication following this link: http://cosmosdb.lambrate.inaf.it/VVDS-SINFONI.

A relationship between specific star formation rate and metallicity gradient within z ∼ 1 galaxies from KMOS-HiZELS

“…We note from Stott et al (2013a) that galaxies on the main sequence at any redshift have a low major merger fraction (∼10 per cent) whereas those in the starburst region may have a merger fraction of ∼50 per cent. Environmental classifications for isolated and potentially interacting galaxies exist for the Queyrel et al (2012) sample (see Epinat et al 2012). From this, we find that the interacting galaxies do have a higher biweight average metallicity gradient of 0.028 ± 0.010 dex kpc −1 than the isolated galaxies, which have an average gradient of 0.003 ± 0.008 dex kpc −1 , but this is only an ∼2σ difference.…”

“…3 is the metallicity gradient plotted against sSFR for which we do see a trend, which is strengthened when our data are combined with those of Queyrel et al (2012), Jones et al (2013) and Swinbank et al (2012). As above, we perform a fit to this combined high-redshift sample, which has the form Z r = c log(sSFR) + d, where c = 0.020 ± 0.007 and d = 0.18 ± 0.07 (i.e.…”

“…We note that the Jones et al (2013) data are for gravitationally lensed galaxies, and therefore the metallicity gradients may be subject to the uncertainties in reconstructing the galaxy images, although they have the advantage of being at high spatial resolution. We perform an outlier-resistant linear regression to the combined high-redshift sample of KMOS-HiZELS, Queyrel et al (2012), Jones et al (2013) and Swinbank et al (2012), which has the form Z r = a log(M ) + b, where a = −0.022 ± 0.009 and b = 0.22 ± 0.03 (i.e. the slope is 2.4σ from being flat).…”

“…Left: the metallicity gradient plotted against stellar mass for the KMOS-HiZELS sample. We include the high-redshift samples of Swinbank et al (2012) and Queyrel et al (2012), and the local samples of normal and merging systems from Rupke et al (2010b). We also plot two galaxies from the lensing sample of Jones et al (2013).…”

“…Shields 1974;McCall, Rybski & Shields 1985;Vila-Costas & Edmunds 1992;Zaritsky, Kennicutt & Huchra 1994;Garnett et al 1997;van Zee et al 1998;Bresolin, Kennicutt & Ryan-Weber 2012;Sánchez et al 2012Sánchez et al , 2014 and see the review by Henry & Worthey 1999). At high redshift, detailed observations become more challenging and the results more contradictory with some authors finding abundance gradients that are consistent with being flat or negative (Swinbank et al 2012;Jones et al 2013) and others seeing evidence for positive gradients Queyrel et al 2012).…”