We explore the redshift evolution of the specific star formation rate (SSFR) for galaxies of different stellar mass by drawing on a deep 3.6 µm-selected sample of > 10 5 galaxies in the 2 deg 2 COSMOS field. The average star formation rate (SFR) for sub-sets of these galaxies is estimated with stacked 1.4 GHz radio continuum emission. We separately consider the total sample and a subset of galaxies that shows evidence for substantive recent star formation in the rest-frame optical spectral energy distributions. At redshifts 0.2 < z < 3 both populations show a strong and mass-independent decrease in their SSFR towards the present epoch. It is best described by a power-law (1 + z) n , where n ∼ 4.3 for all galaxies and n ∼ 3.5 for star forming (SF) sources. The decrease appears to have started at z > 2, at least for high-mass (M * 4 × 10 10 M ⊙ ) systems where our conclusions are most robust. Our data show that there is a tight correlation with power-law dependence, SSFR ∝ M * β , between SSFR and stellar mass at all epochs. The relation tends to flatten below M * ≈ 10 10 M ⊙ if quiescent galaxies are included; if they are excluded from the analysis a shallow index β SFG ≈ −0.4 fits the correlation. On average, higher mass objects always have lower SSFRs, also among SF galaxies. At z > 1.5 there is tentative evidence for an upper threshold in SSFR that an average galaxy cannot exceed, possibly due to gravitationally limited molecular gas accretion. It is suggested by a flattening of the SSFR-M * relation (also for SF sources), but affects massive (> 10 10 M ⊙ ) galaxies only at the highest redshifts. Since z = 1.5 there thus is no direct evidence that galaxies of higher mass experience a more rapid waning of their SSFR than lower mass SF systems. In this sense, the data rule out any strong 'downsizing' in the SSFR. We combine our results with recent measurements of the galaxy (stellar) mass function in order to determine the characteristic mass of a SF galaxy: we find that since z ∼ 3 the majority of all new stars were always formed in galaxies of M * = 10 10.6±0.4 M ⊙ . In this sense, too, there is no 'downsizing'. Finally, our analysis constitutes the most extensive SFR density determination with a single technique out to z = 3. Recent Herschel results are consistent with our results, but rely on far smaller samples.Note. -Median stacking-based average 1.4 GHz radio flux densities and derived average quantities for all our bins in mass and redshift for star forming systems within our mass-selected sample. For details see caption of Tab. 2. † Mass bin contains data below the limit of mass representativeness and yields an upper limit to the average SFR (see Sec. 2.6 for further details.) ⋆ First mass bin above the limit of representativeness (see Sec. 2.6) which contains a low fraction (< 15 %) of optically faint objects with m AB (i + ) ≥ 25.5 for which the photo-z accuracy is degraded (see Sec. 2.2 for further details). The average SFR measured in this bin might be slightly overestimated towards higher values (se...

We estimate the galaxy stellar mass function and stellar mass density for star-forming and quiescent galaxies with 0.2 < z < 4. We construct a large, deep (K s < 24) sample of 220 000 galaxies selected using the new UltraVISTA DR1 data release. Our analysis is based on precise 30-band photometric redshifts. By comparing these photometric redshifts with 10,800 spectroscopic redshifts from the zCOSMOS bright and faint surveys, we find a precision of σ Δz/(1+z) = 0.008 at i + < 22.5 and σ Δz/(1+z) = 0.03 at 1.5 < z < 4. We derive the stellar mass function and correct for the Eddington bias. We find a mass-dependent evolution of the global and starforming populations, with the low-mass end of the mass functions evolving more rapidly than the high-mass end. This mass-dependent evolution is a direct consequence of the star formation being "quenched" in galaxies more massive than M 10 10.7−10.9 M . For the mass function of the quiescent galaxies, we do not find any significant evolution of the high-mass end at z < 1; however we observe a clear flattening of the faint-end slope. From z ∼ 3 to z ∼ 1, the density of quiescent galaxies increases over the entire mass range. Their comoving stellar mass density increases by 1.6 dex between z ∼ 3 and z ∼ 1 and by less than 0.2 dex at z < 1. We infer the star formation history from the mass density evolution. This inferred star formation history is in excellent agreement with instantaneous star formation rate measurements at z < 1.5, while we find differences of 0.2 dex at z > 1.5 consistent with the expected uncertainties. We also develop a new method to infer the specific star formation rate from the mass function of star-forming galaxies. We find that the specific star formation rate of 10 10−10.5 M galaxies increases continuously in the redshift range 1 < z < 4. Finally, we compare our results with a semi-analytical model and find that these models overestimate the density of low mass quiescent galaxies by an order of magnitude, while the density of low-mass star-forming galaxies is successfully reproduced.Key words. galaxies: distances and redshifts -galaxies: evolution -galaxies: formation -galaxies: star formationgalaxies: stellar content Based on data products from observations made with ESO Telescopes at the La Silla Paranal Observatory under ESO programme ID 179.A-2005 and on data products produced by TERAPIX and the Cambridge Astronomy Survey Unit on behalf of the UltraVISTA consortium.Catalogues are only available at the CDS via anonymous ftp to cdsarc.u-strasbg.fr (130.79.128.5) or via

The ALESS survey has followed-up a sample of 122 sub-millimeter sources in the Extended Chandra Deep Field South at 870µm with ALMA, allowing to pinpoint the positions of sub-millimeter galaxies (SMGs) to ∼ 0.3 arcsec and to find their precise counterparts at different wavelengths. This enabled the first compilation of the multi-wavelength spectral energy distributions (SEDs) of a statistically reliable survey of SMGs. In this paper, we present a new calibration of the MAGPHYS SED modelling code that is optimized to fit these ultraviolet-to-radio SEDs of z > 1 star-forming galaxies using an energy balance technique to connect the emission from stellar populations, dust attenuation and dust emission in a physically consistent way. We derive statistically and physically robust estimates of the photometric redshifts and physical parameters (such as stellar masses, dust attenuation, star formation rates, dust masses) for the ALESS SMGs. We find that the ALESS SMGs have a median stellar mass M * = (8.9 ± 0.1) × 10 10 M ⊙ , median star formation rate SFR = 280 ± 70 M ⊙ yr −1 , median overall V -band dust attenuation A V = 1.9 ± 0.2 mag, median dust mass M dust = (5.6 ± 1.0) × 10 8 M ⊙ , and median average dust temperature T dust ≃ 40 K. We find that the average intrinsic spectral energy distribution of the ALESS SMGs resembles that of local ultra-luminous infrared galaxies in the infrared range, but the stellar emission of our average SMG is brighter and bluer, indicating lower dust attenuation, possibly because they are more extended. We explore how the average SEDs vary with different parameters (redshift, sub-millimeter flux, dust attenuation and total infrared luminosity), and we provide a new set of SMG templates that can be used to interpret other SMG observations. To put the ALESS SMGs into context, we compare their stellar masses and star formation rates with those of less actively star-forming galaxies at the same redshifts. We find that, at z ≃ 2, about half of the SMGs lie above the star-forming main sequence (with star formation rates three times larger than normal galaxies of the same stellar mass), while half are consistent with being at the high-mass end of the main sequence. At higher redshifts (z ≃ 3.5), the SMGs tend to have higher star formation rates and stellar masses, but the fraction of SMGs that lie significantly above the main sequence decreases to less than a third.