Preview

PDF, English Download (35MB) | Terms of use

Abstract

In the standard picture of structure formation, the first massive galaxies form in the highest peaks of the density field, which are the cores of massive proto-clusters. Such structures must be exceedingly rare, and thus hard to find. Luminous quasars (QSOs) at z ~ 4 are the most strongly clustered population in the Universe and this large auto-correlation demands that they reside in massive dark matter halos, associated with large overdensities of galaxies. This imply a strong QSO-galaxy cross-correlation for luminous QSOs at z ~ 4. In order to put the first observational constrain on the QSO-galaxy clustering properties, I present the measurement of the cross-correlation function between QSOs and both Lyman break galaxies (LBGs) and Lyman alpha emitters (LAEs) at z ~ 4. I present VLT/FORS1 observations of 6 luminous QSOs fields. Using a novel narrow band filter technique I select LBGs associated with each quasar in a narrow redshift range of Δz ~ 0.2. I measure the QSO-LBG cross correlation function on scales of 0.1 < R < 10 h-1 Mpc (comoving), which is well fitted by a power law form with a correlation length r0 = 9.91+3.28-1.79 h-1 Mpc and a slope gamma = 2.05+0.20-0.46. This is in agreement with the theoretical expected clustering computed from the individual QSO and LBGs auto-correlation, assuming a linear bias model. I also measure the auto-correlation of LBGs in the QSO fields, which shows a stronger clustering in comparison with LBGs in random fields. The new technique used for the detection of LBGs is efficient in selecting them in a thin redshift slice, but this shows important shortcomings whereby this result should be carefully considered. Additionally, I present VLT/FORS2 observations of 14 luminous QSO fields, designed to search for LAEs in their environments. I find that QSOs and LAEs are not correlated at z ~ 4, and the mean LAEs number density in our fields is consistent with the expected number density in random fields. This could mean either that the QSO auto-correlation length at z ~ 4 is overestimated, or that LAEs preferentially avoid QSO environments on ≲10 Mpc h-1 scales.