We present a study of the evolution of several classes of Mg ii absorbers, and their corresponding Fe ii absorption, over a large fraction of cosmic history: 2.3–8.7 Gyr from the big bang. Our sample consists of 87 strong ( Å) Mg ii absorbers, with redshifts 0.2 < z < 2.5, measured in 81 quasar spectra obtained from the Very Large Telescope/Ultraviolet and Visual Echelle Spectrograph archives of high-resolution spectra (R ∼ 45 000). No evolutionary trend in is found for moderately strong Mg ii absorbers ( Å). However, at lower redshifts we find an absence of very strong Mg ii absorbers (those with Å) with small ratios of equivalent widths of Fe ii to Mg ii. At high redshifts, very strong Mg ii absorbers with both small and large values are present. We compare our findings to a sample of 100 weak Mg ii absorbers ( Å) found in the same quasar spectra by Narayanan et al.
The main effect driving the evolution of very strong Mg ii systems is the difference between the kinematic profiles at low and high redshift. At high redshift, we observe that, among the very strong Mg ii absorbers, all of the systems with small ratios of have relatively large velocity spreads, resulting in less saturated profiles. At low redshift, such kinematically spread systems are absent, and both Fe ii and Mg ii are saturated, leading to values that are all close to 1. The high redshift, small systems could correspond to sub-damped Lyman α systems, many of which have large velocity spreads and are possibly linked to superwinds in star-forming galaxies. In addition to the change in saturation due to kinematic evolution, the smaller values could be due to a lower abundance of Fe at high redshifts, which would indicate relatively early stages of star formation in those environments.