The subject of the evolution of the active galactic nucleus (AGN) population began in the late 1960s with Schmidt's (1968, 1970) discoveries that the space densities of both radio and optically selected quasars increased significantly with redshift. The effect was so strong that it was detectable in samples as small as 20 objects. He developed and applied the V / Vm test for analyzing the space distribution in his samples and showed that there was a strong evolution in the space density of quasars toward higher redshift, increasing by more than a factor of 100 from redshift 0 to 2. This was a striking and unexpected result that posed a question that is still crucial today -- What causes the sharp decline since -- z = 2?
In this article I will review and discuss the following subjects:
The techniques used to discover quasars and AGNs, their selection effects, and the surveys used to study the evolution of the AGN population.
The general picture of evolution up to 1995, when the first well-defined, quantitative surveys of the evolution of high-redshift quasars were published.
The current status of major optical surveys such as 2dF and SDSS.
Radio and X-ray surveys and how they are critical to understanding AGN evolution.
The relation of AGNs to their host galaxies and how studies of massive black holes in spheroids provide constraints on AGN evolution.
Current research problems, such as measuring the quasar luminosity function at high redshift and faint magnitudes; relating observed to physical evolution; the framework for connecting observations, accretion processes, and the growth of black hole masses; and how to estimate black hole masses.
Before proceeding, let us define and discuss terms used in this article to aid the clarity of the presentation.
An AGN is one not powered by normal stellar processes, although active star formation may be occurring in the vicinity. The working hypothesis is that AGNs contain massive black holes and are powered by accretion processes. Their luminosities range from as low as MB = - 9 mag to as high as MB = - 30 mag (LX = 1038 to 1048 erg s-1). Quasars are the high-luminosity (MB < - 23 mag, LX > 1044 erg s-1) members of the AGN family.
Traditionally, evolution of AGNs or quasars has meant the evolution with redshift of their luminosity function or space density (which is the integral of the luminosity function over some range of luminosities). However, evolution can also refer to changes with redshift of the spectral energy distribution (SED) or the emission-line spectra of AGNs. In general, observed evolution will refer to changes with redshift of any observed property of AGNs.
Ultimately, we wish to map and understand the physical evolution of AGNs, by which we mean how their central black holes form and grow with cosmic epoch and how their accretion processes and rates, which determine the luminosities and SEDs we observe from AGNs, evolve with cosmic epoch. The discovery of the ubiquity of black holes in the spheroids of nearby galaxies makes us realize that the physical evolution of AGNs is closely connected with and is an important part of the larger subject of how galaxies in general form and evolve. It appears that virtually every spheroidal system went through an AGN phase at some time in its history -- thus the subject of our meeting: "The Coevolution of Black Holes and Galaxies."
However, the persistent question of how many AGNs are hidden because of weak emission lines, obscuration by dust, or absorption in X-rays has continued to impede progress in the mapping of the observational evolution of AGNs and must be addressed in any attempt to determine the properties of the overall AGN population. Fortunately, the advent of powerful new space observatories such as Chandra and XMM-Newton, in conjunction with sensitive radio and infrared surveys, provides new tools for attacking this problem, as will be addressed below.
At the same time, the formulation and application of the appropriate observational definitions of AGNs continue to be critical issues in current research, especially for low-luminosity objects, which can be hard to find within the glare of their host galaxy or to separate from normal stars. For example, the work of Ho (2003) suggests that some AGNs may have X-ray luminosities down to 1036 erg s-1, or less than stellar X-ray sources.
If we are really to understand the global population of AGNs and their relation to galaxies, these problems must be solved. This will be one of the themes to be developed in this article.