We investigate the relation between star formation rate (SFR) and stellar mass (M), i.e. the Main Sequence (MS) relation of star-forming galaxies, at 1.3 ≤ z < 6 in the first four HST Frontier Fields, based on rest-frame UV observations. Gravitational lensing combined with deep HST observations allows us to extend the analysis of the MS down to log M/M ⊙ ∼ 7.5 at z 4 and log M/M ⊙ ∼ 8 at higher redshifts, a factor of ∼10 below most previous results. We perform an accurate simulation to take into account the effect of observational uncertainties and correct for the Eddington bias. This step allows us to reliably measure the MS and in particular its slope. While the normalization increases with redshift, we fit an unevolving and approximately linear slope. We nicely extend to lower masses the results of brighter surveys. Thanks to the large dynamic range in mass and by making use of the simulation, we analyzed any possible mass dependence of the dispersion around the MS. We find tentative evidence that the scatter decreases with increasing mass, suggesting larger variety of star formation histories in low mass galaxies. This trend agrees with theoretical predictions, and is explained as either a consequence of the smaller number of progenitors of low mass galaxies in a hierarchical scenario and/or of the efficient but intermittent stellar feedback processes in low mass halos. Finally, we observe an increase in the SFR per unit stellar mass with redshift milder than predicted by theoretical models, implying a still incomplete understanding of the processes responsible for galaxy growth.

We reduce and analyse the available James Webb Space Telescope (JWST) ERO and ERS NIRCam imaging (SMACS0723, GLASS, CEERS) in combination with the latest deep ground-based near-infrared imaging in the COSMOS field (provided by UltraVISTA DR5) to produce a new measurement of the evolving galaxy UV luminosity function (LF) over the redshift range z = 8 − 15. This yields a new estimate of the evolution of UV luminosity density (ρUV), and hence cosmic star-formation rate density (ρSFR) out to within &lt;300 Myr of the Big Bang. Our results confirm that the high-redshift LF is best described by a double power-law (rather than a Schechter) function up to z ∼ 10, and that the LF and the resulting derived ρUV (and thus ρSFR), continues to decline gradually and steadily up to z ∼ 15 (as anticipated from previous studies which analysed the pre-existing data in a consistent manner to this study). We provide details of the 61 high-redshift galaxy candidates, 47 of which are new, that have enabled this new analysis. Our sample contains 6 galaxies at z ≥ 12, one of which appears to set a new redshift record as an apparently robust galaxy candidate at z ≃ 16.4, the properties of which we therefore consider in detail. The advances presented here emphasize the importance of achieving high dynamic range in studies of early galaxy evolution, and re-affirm the enormous potential of forthcoming larger JWST programmes to transform our understanding of the young Universe.

We present the results of a study utilising ultra-deep, rest-frame UV, spectroscopy to quantify the relationship between stellar mass and stellar metallicity for 681 starforming galaxies at 2.5 < z < 5.0 ( z = 3.5 ± 0.6) drawn from the VANDELS survey. Via a comparison with high-resolution stellar population synthesis models, we determine stellar metallicities (Z * , here a proxy for the iron abundance) for a set of high signal-to-noise ratio composite spectra formed from subsamples selected by mass and redshift. Across the stellar mass range 8.5 < log( M * /M ) < 10.2 we find a strong correlation between stellar metallicity (Z * /Z ) and stellar mass, with stellar metallicity monotonically increasing from Z * /Z < 0.09 at M * = 3.2 × 10 8 M to Z * /Z = 0.27 at M * = 1.7 × 10 10 M . In contrast, at a given stellar mass, we find no evidence for significant metallicity evolution across the redshift range of our sample. However, comparing our results to the z = 0 stellar mass-metallicity relation for star-forming galaxies, we find that the z = 3.5 relation is consistent with being shifted to lower metallicities by 0.6 dex at all stellar masses. Contrasting our derived stellar metallicities with estimates of the gas-phase metallicities of galaxies at similar redshifts and stellar masses, we find evidence for enhanced O/Fe ratios in z 2.5 star-forming galaxies of the order (O/Fe) 1.8 × (O/Fe) . Finally, by comparing our results to the predictions of three cosmological simulations, we find that the z = 3.5 stellar mass-metallicity relation is consistent with current predictions for how outflow strength scales with galaxy stellar mass. This conclusion is supported by an analysis of one-zone analytic chemical evolution models, and suggests that the mass loading parameter (η =Ṁ outflow /M * ) scales as η ∝ M β * with β −0.4.