Characterizing the Warm-Hot IGM at High Redshift: A High Resolution Survey for O VI at z = 2.5

Abstract

We describe a survey for OVI absorption in the spectra of 5 high redshift quasars (2.2 < z < 2.8). We identify 12 cosmological systems, and 6 systems that are either ejected from the background QSO or affected by its local radiation field. Almost all of the intergalactic OVI is associated with strong Ly-a absorption (N_HI > 10^15.2 cm^-2), as well as absorption from CIV and often lower ionization species. The absorbing regions are conservatively constrained to have L<=200 kpc and rho/rho_bar >= 2.5, with actual values probably closer to L ~ 60 kpc and rho/rho_bar ~ 10-30. They also have two distinct gas phases: one which produces photoionized CIV and SiIV at T ~ 30,000 K, and a second which is seen only in OVI. The OVI temperature is difficult to constrain due to uncertainty in the amount of nonthermal line broadening, but it does appear that this gas is hotter than the CIV/SiIV phase and could support collisional OVI production. The OVI is strongly clustered on velocity scales of dv=100-300 km/s, with weaker signal extending to dv = 750 km/s. The power-law slope of the correlation function resembles that of local galaxy and cluster surveys, with a comoving correlation length of 11h_{65}^-1 Mpc. The average Oxygen abundance of the OVI systems is [O/H]>-1.5, about 10 times higher than the level observed in the general IGM. Two OVI production mechanisms are considered: shock heating of gas falling onto existing structure, and expulsion of material by galactic winds. Simulations of infall models tend to overproduce OVI lines by a factor of ~10, though this may result from numerical limitations. Known galaxy populations such as the Lyman break objects could produce the observed amount of OVI if they drive winds to distances of R ~ 50 kpc.