2017MNRAS.469.2959K

Starting from a summary of detection statistics of our recent X-shooter campaign, we review the major surveys, both space and ground based, for emission counterparts of high-redshift damped Ly α absorbers (DLAs) carried out since the first detection 25 yr ago. We show that the detection rates of all surveys are precisely reproduced by a simple model in which the metallicity and luminosity of the galaxy associated to the DLA follow a relation of the form, M_UV = -5 x ([M/H] + 0.3) - 20.8, and the DLA cross-section follows a relation of the form σ_ DLA_ ∝L^0.8. Specifically, our spectroscopic campaign consists of 11 DLAs pre-selected based on their equivalent width of Si II λ1526 to have a metallicity higher than [Si/H] > -1. The targets have been observed with the X-shooter spectrograph at the Very Large Telescope to search for emission lines around the quasars. We observe a high detection rate of 64 per cent (7/11), significantly higher than the typical ∼10 per cent for random, H I-selected DLA samples. We use the aforementioned model, to simulate the results of our survey together with a range of previous surveys: spectral stacking, direct imaging (using the 'double DLA' technique), long-slit spectroscopy, and integral field spectroscopy. Based on our model results, we are able to reconcile all results. Some tension is observed between model and data when looking at predictions of Ly α emission for individual targets. However, the object-to-object variations are most likely a result of the significant scatter in the underlying scaling relations as well as uncertainties in the amount of dust which affects the emission.