The Evolution of Quasar CIV and SiIV Broad Absorption Lines Over Multi-Year Time Scales

TL;DR

This study examines the multi-year variability of quasar broad absorption lines, revealing complex, non-monotonic changes, constraining BAL lifetimes to over 30 years, and exploring the relationship between CIV and SiIV absorption variations.

Contribution

It provides the first multi-year analysis of BAL variability in quasars, highlighting the complex patterns and constraining the BAL lifetime to over 30 years.

Findings

BAL variability increases with time between observations.

BALs do not show monotonic strengthening or weakening.

BAL lifetime is constrained to be greater than approximately 30 years.

Abstract

We investigate the variability of CIV 1549A broad absorption line (BAL) troughs over rest-frame time scales of up to ~7 yr in 14 quasars at redshifts z>2.1. For 9 sources at sufficiently high redshift, we also compare CIV and SiIV 1400A absorption variation. We compare shorter- and longer-term variability using spectra from up to four different epochs per source and find complex patterns of variation in the sample overall. The scatter in the change of absorption equivalent width (EW), Delta EW, increases with the time between observations. BALs do not, in general, strengthen or weaken monotonically, and variation observed over shorter (<months) time scales is not predictive of multi-year variation. We find no evidence for asymmetry in the distribution of Delta EW that would indicate that BALs form and decay on different time scales, and we constrain the typical BAL lifetime to be >~30 yr. The BAL absorption for one source, LBQS 0022+0150, has weakened and may now be classified as a mini-BAL. Another source, 1235+1453, shows evidence of variable, blue continuum emission that is relatively unabsorbed by the BAL outflow. CIV and SiIV BAL shape changes are related in at least some sources. Given their high velocities, BAL outflows apparently traverse large spatial regions and may interact with parsec-scale structures such as an obscuring torus. Assuming BAL outflows are launched from a rotating accretion disk, notable azimuthal symmetry is required in the outflow to explain the relatively small changes observed in velocity structure over times up to 7 yr.