The star - formation rate of Ca II and damped Lyman - α absorbers at 0 . 4 < z < 1

Using stacked Sloan Digital Sky Survey spectra, we present the detection of [O II]λλ3727,3730 nebular emission from the galaxies hosting Ca II absorption line systems and galaxies hosting Mg II-selected damped Lyman-α (DLA) absorbers. Both samples of absorbers, 345 Ca II systems and 3461 Mg II-selected DLA systems, span the redshift interval 0.4 6 zabs < 1.3; all of the former and half the latter sample are expected to be bonafide DLAs. The measured star formation rate (SFR) per absorber from light falling within the SDSS fibre apertures (corresponding to physical radii of 6−9h kpc) is 0.11−0.14 M⊙yr for the Mg II-selected DLAs and 0.11 − 0.48 M⊙yr−1 for the Ca II-absorbers. These results represent the first estimates of the average SFR in an absorption-selected galaxy population from the direct detection of nebular emission. Adopting the currently favoured model in which DLAs are large, with radii > ∼ 9h kpc, and assuming no attenuation of the [O II] emission by dust, leads us to conclude that the SFR per unit area of Mg II-selected DLAs falls an order of magnitude below the predictions of the Schmidt law, which relates the SFR to the H I column density at z∼0. While DLA sightlines are known to contain little dust, the unknown geometry of the dust distribution in the galaxies causes the main uncertainty in our results. The contribution of both DLA and Ca II absorbers to the total observed star formation rate density, ρ̇, in the redshift range 0.4 < z < 1.3, is small, < ∼ 10% and < ∼ 3% respectively. The result contrasts with the conclusions of Hopkins et al. that DLA absorbers can account for the majority of the total observed ρ̇ in the same redshift range. The disagreement is a direct consequence of the much lower SFR per unit area we observe than predicted by the Schmidt law. Our results effectively rule out a picture in which DLA absorbers are the sites in which a large fraction of the total ρ̇ at redshifts z < ∼ 1 occurs.