Search

We study the physical properties (size, stellar mass, luminosity, star formation rate) and scaling relations for a sample of 166 star-forming clumps with redshift $z \sim 2-6.2$. They are magnified by the Hubble Frontier Field galaxy cluster MACS~J0416 and have robust lensing magnification ($2\lesssim \mu \lesssim 82$) computed by using our high-precision lens model, based on 182 multiple images. Our sample extends by $\sim 3$ times the number of spectroscopically-confirmed lensed clumps at $z \gtrsim 2$. We identify clumps in ultraviolet continuum images and find that, whenever the effective spatial resolution (enhanced by gravitational lensing) increases, they fragment into smaller entities, likely reflecting the hierarchically-organized nature of star formation. Kpc-scale clumps, most commonly observed in field, are not found in our sample. The physical properties of our sample extend the parameter space typically probed by $z \gtrsim 1$ field observations and simulations, by populating the low mass (M$_\star \lesssim 10^7$ M$_\odot$), low star formation rate (SFR $\lesssim 0.5$ M$_\odot$ yr$^{-1}$), and small size (R$_\mathrm{eff} \lesssim 100$ pc) regime. The new domain probed by our study approaches the regime of compact stellar complexes and star clusters. In the mass-size plane, our sample spans the region between galaxies and globular clusters, with a few clumps in the region populated by young star clusters and globular-clusters. For the bulk of our sample, we measure star-formation rates which are higher than those observed locally in compact stellar systems, indicating different conditions for star formation at high redshift than in the local Universe.

We empirically test the relation between the SFR(LIR) derived from the infrared luminosity, LIR, and the SFR(Ha) derived from the Ha emission line luminosity using simple conversion relations. We use a sample of 474 galaxies at z = 0.06 - 0.46 with both Ha detection (from 20k zCOSMOS survey) and new far-IR Herschel data (100 and 160 {\mu}m). We derive SFR(Ha) from the Ha extinction corrected emission line luminosity. We find a very clear trend between E(B - V) and LIR that allows to estimate extinction values for each galaxy even if the Ha emission line measurement is not reliable. We calculate the LIR by integrating from 8 up to 1000 {\mu}m the SED that is best fitting our data. We compare SFR(Ha) with the SFR(LIR). We find a very good agreement between the two SFR estimates, with a slope of m = 1.01 \pm 0.03 in the SFR(LIR) vs SFR(Ha) diagram, a normalization constant of a = -0.08 \pm 0.03 and a dispersion of sigma = 0.28 dex.We study the effect of some intrinsic properties of the galaxies in the SFR(LIR)-SFR(Ha) relation, such as the redshift, the mass, the SSFR or the metallicity. The metallicity is the parameter that affects most the SFR comparison. The mean ratio of the two SFR estimators log[SFR(LIR)/SFR(Ha)] varies by approx. 0.6 dex from metal-poor to metal-rich galaxies (8.1 < log(O/H) + 12 < 9.2). This effect is consistent with the prediction of a theoretical model for the dust evolution in spiral galaxies. Considering different morphological types, we find a very good agreement between the two SFR indicators for the Sa, Sb and Sc morphologically classified galaxies, both in slope and normalization. For the Sd, irregular sample (Sd/Irr), the formal best-fit slope becomes much steeper (m = 1.62 \pm 0.43), but it is still consistent with 1 at the 1.5 sigma level, because of the reduced statistics of this sub-sample.