On the Redshift Evolution of MgII Absorption Systems

We use a halo occupation approach to connect MgII absorbers to dark matter halos as a function of redshift. Using the model constructed in Tinker & Chen (2008), we parameterize the conditional probability of an absorber of equivalent width Wr being produced by a halo of mass M_h at a given redshift, P(Wr|M_h,z). We constrain the free parameters of the model by matching the observed statistics of MgII absorbers: the frequency function f(Wr), the redshift evolution n(z), and the clustering bias b(Wr). The redshift evolution of Wr>1 A absorbers increases from z=0.4 to z=2, while the total halo cross section decreases monotonically with redshift. This discrepancy can only be explained if the gaseous halos evolve with respect to their host halos. We make predictions for the clustering bias of absorbers as a function of redshift under different evolutionary scenarios, eg, the gas cross section per halo evolves or the halo mass scale of absorbers changes. We demonstrate that the relative contribution of these scenarios may be constrained by measurements of absorber clustering at z>1 and z~0.1. If we further assume a redshift-independent mass scale for efficient shock heating of halo gas of Mcrit = 10^{11.5} Msol/h, absorber evolution is predominantly caused by a changing halo mass scale of absorbers. Our model predicts that strong absorbers always arise in ~Mcrit halos, independent of redshift, but the mass scale of weak absorbers decreases by 2 dex from 0<z<2. Thus, the measured anti-correlation of clustering bias and Wr should flatten by z~1.5.