The VLT LBG Redshift Survey – III. The clustering and dynamics of Lyman-break galaxies at z ∼ 3★

We present a catalogue of 2135 galaxy redshifts from the VLT LBG Redshift Survey (VLRS), a spectroscopic survey of z 3 galaxies in wide fields centred on background quasi-stellar objects. We have used deep optical imaging to select galaxies via the Lyman-break technique. Spectroscopy of the Lyman-break galaxies (LBGs) was then made using the Very Large Telescope (VLT) Visible Multi-Object Spectrograph (VIMOS) instrument, giving a mean redshift of z = 2.79. We analyse the clustering properties of the VLRS sample and also of the VLRS sample combined with the smaller area Keck-based survey of Steidel et al. From the semiprojected correlation function, w p ( ), for the VLRS and combined surveys, we find that the results are well fit with a single power-law model, with clustering scale lengths of r 0 = 3.46 0.41 and 3.83 0.24 h -1 Mpc, respectively. We note that the corresponding combined (r) slope is flatter than for local galaxies at = 1.5-1.6 rather than = 1.8. This flat slope is confirmed by the z-space correlation function, (s), and in the range 10 < s < 100 h -1 Mpc the VLRS shows an 2.5 excess over the cold dark matter ( CDM) linear prediction. This excess may be consistent with recent evidence for non-Gaussianity in clustering results at z 1. We then analyse the LBG z-space distortions using the 2D correlation function, ( , ), finding for the combined sample a large-scale infall parameter of = 0.38 0.19 and a velocity dispersion of w 2 z = 420 +140 -160 km s -1 . Based on our measured , we are able to determine the gravitational growth rate, finding a value of f(z = 3) = 0.99 0.50 (or f 8 = 0.26 0.13), which is the highest redshift measurement of the growth rate via galaxy clustering and is consistent with CDM. Finally, we constrain the mean halo mass for the LBG population, finding that the VLRS and combined sample suggest mean halo masses of log(M DM /M ) = 11.57 0.15 and 11.73 0.07, respectively.