We present a study of the evolution of several classes of MgII absorbers, and their corresponding FeII absorption, over a large fraction of cosmic history: 2.3 to 8.7 Gyrs from the Big Bang. Our sample consists of 87 strong (Wr(MgII)>0.3 A) MgII absorbers, with redshifts 0.2<z<2.5, measured in 81 quasar spectra obtained from the Very Large Telescope(VLT)/Ultraviolet and Visual Echelle Spectrograph(UVES) archives of high-resolution spectra (R \sim 45,000). No evolutionary trend in Wr(FeII)/Wr(MgII) is found for moderately strong MgII absorbers (0.3<Wr(MgII)<1.0 A). However, at lower z we find an absence of very strong MgII absorbers (those with Wr(MgII)>1 A) with small ratios of equivalent widths of FeII to MgII. At high z, very strong MgII absorbers with both small and large Wr(FeII)/Wr(MgII) values are present. We compare our findings to a sample of 100 weak MgII absorbers (Wr(MgII)<0.3 A) found in the same quasar spectra by Narayanan et al. (2007). The main effect driving the evolution of very strong MgII systems is the difference between the kinematic profiles at low and high redshifts. At high z, we observe that, among the very strong MgII absorbers, all of the systems with small ratios of Wr(FeII)/Wr(MgII) have relatively large velocity spreads, resulting in less saturated profiles. At low z, such kinematically spread systems are absent, and both FeII and MgII are saturated, leading to Wr(FeII)/Wr(MgII) values that are all close to 1. The high redshift, small Wr(FeII)/Wr(MgII) systems could correspond to sub-DLA 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 Wr(FeII)/Wr(MgII) values could be due to a lower abundance of Fe at high z, which would indicate relatively early stages of star formation in those environments.

Several neutral species (MgI, SiI, CaI, FeI) have been detected in a weak MgII absorption line system (W_r(2796)~0.15 Angstroms) at z~0.45 along the sightline toward HE0001-2340. These observations require extreme physical conditions, as noted in D'Odorico (2007). We place further constraints on the properties of this system by running a wide grid of photoionization models, determining that the absorbing cloud that produces the neutral absorption is extremely dense (~100-1000/cm^3), cold (<100 K), and has significant molecular content (~72-94%). Structures of this size and temperature have been detected in Milky Way CO surveys, and have been predicted in hydrodynamic simulations of turbulent gas. In order to explain the observed line profiles in all neutral and singly ionized chemical transitions, the lines must suffer from unresolved saturation and/or the absorber must partially cover the broad emission line region of the background quasar. In addition to this highly unusual cloud, three other ordinary weak MgII clouds (within densities of ~0.005/cm^3 and temperatures of ~10000K) lie within 500 km/s along the same sightline. We suggest that the "bare molecular cloud", which appears to reside outside of a galaxy disk, may have had in situ star formation and may evolve into an ordinary weak MgII absorbing cloud.

We consider the evolution of strong (W_r(2796) > 0.3A) MgII absorbers, most of which are closely related to luminous galaxies. Using 20 high resolution quasar spectra from the VLT/UVES public archive, we examine 33 strong MgII absorbers in the redshift range 0.3 < z < 2.5. We compare and supplement this sample with 23 strong MgII absorbers at 0.4 < z < 1.4 observed previously with HIRES/Keck. We find that neither equivalent width nor kinematic spread (the optical depth weighted second moment of velocity) of MgII2796 evolve. However, the kinematic spread is sensitive to the highest velocity component, and therefore not as sensitive to additional weak components at intermediate velocities relative to the profile center. The fraction of absorbing pixels within the full velocity range of the system does show a trend of decreasing with decreasing redshift. Most high redshift systems (14/20) exhibit absorption over the entire system velocity range, which differs from the result for low redshift systems (18/36) at the 95% level. This leads to a smaller number of separate subsystems for high redshift systems because weak absorping components tend to connect the stronger regions of absorption. We hypothesize that low redshift MgII profiles are more likely to represent well formed galaxies, many of which have kinematics consistent with a disk/halo structure. High redshift MgII profiles are more likely to show evidence of complex protogalactic structures, with multiple accretion or outflow events. Although these results are derived from measurements of gas kinematics, they are consistent with hierarchical galaxy formation evidenced by deep galaxy surveys.