The star formation rate density and dust attenuation evolution over 12 Gyr with the VVDS surveys

Abstract

Aims. We investigate the global galaxy evolution over ∼12 Gyr (0.05 ≤ z ≤ 4.5), from the far ultraviolet (FUV) luminosity function (LF), luminosity density (LD), and star formation rate density (SFRD), using the VIMOS-VLT Deep Survey (VVDS), a single deep galaxy redshift survey with a well controlled selection function. Methods. We combine the VVDS Deep (17.5 ≤ I AB ≤ 24.0) and Ultra-Deep (23.00 ≤ i AB ≤ 24.75) redshift surveys, totalizing ∼11 000 galaxies, to estimate the rest-frame FUV LF and LD, using a wide wavelength range of deep photometry (337 < λ < 2310 nm). We extract the dust attenuation of the FUV radiation, embedded in the well-constrained spectral energy distributions. We then derive the dust-corrected SFRD.Results. We find a constant and flat faint-end slope α in the FUV LF at z < 1.7. At z > 1.7, we set α steepening with (1+z). The absolute magnitude M * FUV steadily brightens in the entire range 0 < z < 4.5, and at z > 2 it is on average brighter than in the literature, while φ * is on average smaller. The evolution of our total LD shows a peak at z 2, clearly present also when considering all sources of uncertainty. The SFRD history peaks as well at z 2. It first steadily rises by a factor of ∼6 during 2 Gyr (from z = 4.5 to z = 2), and then decreases by a factor of ∼12 during 10 Gyr down to z = 0.05. This peak is mainly produced by a similar peak within the population of galaxies with −21.5 ≤ M FUV ≤ −19.5. As times goes by, the total SFRD is dominated by fainter and fainter galaxies. The mean dust attenuation of the global galaxy population rises fast by 1 mag during 2 Gyr from z 4.5 to z ∼ 2, reaches slowly its maximum at z 1 (A FUV 2.2 mag), and then decreases by 1.1 mag during 7 Gyr down to z 0. Conclusions. We have derived the cosmic SFRD history and the total dust amount in galaxies over a continuous period of ∼12 Gyr, using a single homogeneous spectroscopic redshift sample. The presence of a clear peak at z 2 and a fast rise at z > 2 of the SFRD is compelling for models of galaxy formation. This peak is mainly produced by bright galaxies (L > ∼ L * z=2 ), requiring that significant gas reservoirs still exist at this epoch and are probably replenished by cold accretion and wet mergers, while feedback or quenching processes are not yet strong enough to lower the SF. The dust attenuation maximum is reached ∼2 Gyr after the SFRD peak, implying a contribution from the intermediate-mass stars to the dust production at z < 2.

The impact of protocluster environments atz= 1.6

“…Combining the SFR of all protocluster galaxies with stellar masses above our mass completeness limit of 10 9.7 M ⊙ results in a total SFR of more than 4000 M ⊙ yr −1 within the 10.2 × 10.2 comoving Mpc area covered by our images. This translates into a global star formation rate density (ρ SFR ) of 1.1 M ⊙ yr −1 Mpc −3 , which is ten times greater than the field density at this redshift (Cucciati et al 2012;Magnelli et al 2013). Thirty percent of the star formation occurs within the massive groups of the protocluster (∼ 1200 M ⊙ yr −1 ), implying that the star formation rate density is ρ SFR = 82 M ⊙ yr −1 Mpc −3 within the groups, which is over 800 times the field density.…”

The impact of galactic properties and environment on the quenching of central and satellite galaxies: a comparison between SDSS, Illustris and L-Galaxies

“…The problem arises because in our sample of passive galaxies, most of the galaxies quenched for the first time several billion years ago, when their properties would have (potentially significantly) different values. The peak of quenching in the Universe appears to be at z ∼ 1 − 2, given the sharp decline in the star formation rate density over this epoch (e.g., Lilly et al 1996;Madau et al 1998;Cucciati et al 2012). Lang et al (2014) probed this redshift range using CANDELS 1 data, finding that the quenched fraction correlates more strongly with bulge mass than total stellar mass or morphology (an equivalent result at high redshifts to the low redshift result of Bluck et al 2014).…”

The evolution of the [O ii], H β and [O iii] emission line luminosity functions over the last nine billions years

“…The magnitude limits are i = 22.5 and 24.0 and the effective areas covered are 5.8 and 0.6 deg 2 for the wide and the deep, respectively. The targets were chosen using the i magnitude from CFHT observations (McCracken et al 2003;Ilbert et al 2005;Cucciati et al 2012). The VVDS collaboration provides the slit-extracted 1D-spectra and the redshift catalogue based on visual inspection of the spectra.…”