We used near-infrared integral field spectroscopic observations from the AMAZE and LSD ESO programs to constrain the metallicity in a sample of 40 star-forming galaxies at 3 < z < 5 (most of which are at z ∼ 3.4). We measured metallicities by exploiting strong emission-line diagnostics. We found that a significant fraction of star-forming galaxies at z ∼ 3.4 deviate from the fundamental metallicity relation (FMR), with a metallicity of up to a factor of ten lower than expected according to the FMR. This deviation does not correlate with the dynamical properties of the galaxy or with the presence of interactions. To investigate the origin of the metallicity deviation in more detail, we also inferred information on the gas content by inverting the Schmidt-Kennicutt relation, assuming that the latter does not evolve out to z ∼ 3.4. In agreement with recent CO observational data, we found that in contrast with the steeply rising trend at 0 < z < 2, the gas fraction in massive galaxies remains constant, with an indication of a marginal decline at 2 < z < 3.5. When combined with the metallicity information, we infer that to explain the low metallicity and gas content in z ∼ 3.4 galaxies, both prominent outflows and massive pristine gas inflows are needed. In ten galaxies we can also spatially resolve the metallicity distribution. We found that the metallicity generally anticorrelates with the distribution of star formation and with the gas surface density. We discuss these findings in terms of pristine gas inflows toward the center, and outflows of metal-rich gas from the center toward the external regions.

We present a SINFONI integral-field kinematical study of 33 galaxies at z ∼ 3 from the AMAZE and LSD projects, which are aimed at studying metallicity and dynamics of high-redshift galaxies. The number of galaxies analyzed in this paper constitutes a significant improvement over existing data in the literature, and this is the first time that a dynamical analysis is obtained for a relatively large sample of galaxies at z ∼ 3. Eleven galaxies show ordered rotational motions (∼30% of the sample). In these cases we estimate dynamical masses by modeling the gas kinematics with rotating disks and exponential mass distributions. We find dynamical masses in the range 2 × 10 9 M −2 × 10 11 M with a mean value of ∼2 × 10 10 M . By comparing observed gas velocity dispersion with what is expected from models, we find that most rotating objects are dynamically "hot", with intrinsic velocity dispersions of ∼90 km s −1 . The median value of the ratio between the maximum disk rotational velocity and the intrinsic velocity dispersion for the rotating objects is 1.6, much lower than observed in local galaxies value (∼10) and slightly lower than the z ∼ 2 value (2-4). Finally we use the maximum rotational velocity from our modeling to build a baryonic Tully-Fisher relation at z ∼ 3. Our measurements indicate that z ∼ 3 galaxies have lower stellar masses (by a factor of ten on average) compared to local galaxies with the same dynamical mass. However, the large observed scatter suggests that the Tully-Fisher relation is not yet "in place" at these early cosmic ages, possibly owing to the young age of galaxies. A smaller dispersion of the Tully-Fisher relation is obtained by taking the velocity dispersion into account with the use of the S 0.5 indicator, suggesting that turbulent motions might play an important dynamical role.

Aims. We aim to constrain the assembly history of high-redshift galaxies and the reliability of UV-based estimates of their physical parameters from an accurate analysis of a unique sample of z ∼ 3 Lyman-break galaxies (LBGs). Methods. We analyse 14 LBGs at z ∼ 2.8-3.8 constituting the only sample where both a spectroscopic measurement of their metallicity and deep IR observations (CANDELS+HUGS survey) are available. Fixing the metallicity of population synthesis models to the observed values, we determine best-fit physical parameters under different assumptions about the star-formation history (SFH) and also consider the effect of nebular emission. For comparison, we determine the UV slope of the objects, and use it to estimate their SFR UV99 by correcting the UV luminosity. Results. A comparison between star-formation rate (SFR) obtained through SED-fitting (SFR fit ) and the SFR UV99 shows that the latter are underestimated by a factor of 2-10, regardless of the assumed SFH. Other SFR indicators (radio, far-IR, X-ray, recombination lines) coherently indicate SFRs a factor of 2-4 larger than SFR UV99 and in closer agreement with SFR fit . This discrepancy is due to the solar metallicity implied by the usual β−A 1600 conversion factor. We propose a refined relation, appropriate for subsolar metallicity LBGs: A 1600 = 5.32 + 1.99 * β. This relation reconciles the dust-corrected UV with the SED-fitting and the other SFR indicators. We show that the fact that z ∼ 3 galaxies have subsolar metallicity implies an upward revision by a factor of ∼1.5-2 of the global SFRD, depending on the assumptions about the age of the stellar populations. We find very young best-fit ages (10-500 Myr) for all our objects. From a careful examination of the uncertainties in the fit and the amplitude of the Balmer break we conclude that there is little evidence of the presence of old stellar population in at least half of the LBGs in our sample, suggesting that these objects are probably caught during a huge star-formation burst, rather than being the result of a smooth evolution.