Identifying active galaxies

Once a new class of object has been identified, we want to ask questions about the space density, luminosity function, and environment of these objects. This requires obtaining a large sample of these sources, and that requires an efficient means of separating them from other objects in the sky.

In the case of AGNs, all of the characteristics mentioned previously can be used to isolate them. Strategies for obtaining large samples of AGNs would thus include (a) radio surveys, (b) searches based on photometric color ("ultraviolet excess"), (c) searches for objects with strong emission lines, and (d) searches for variable sources. Indeed, all of these methods have been used to identify AGN candidates.

As already noted, the first quasars were discovered as a result of radio surveys of the sky. A serious difficulty with this approach was that the angular resolution of single-dish antennas is quite low (typically arcminutes) compared with ground-based resolution of the sky at optical wavelengths (typically around an arcsecond). Even at high Galactic latitude, there are usually multiple candidate objects that might be associated with the radio sources. However, once a few quasars with particularly accurate radio-source positions had been identified, it became clear that the quasars stand out in the U — B color (Figures 5.1 and 5.2), greatly simplifying their identification among the candidates. This led to the further realization that other UV-excess sources at high Galactic latitude might also be quasars, and one might by-pass the laborious radio observations altogether. Searching for quasars by detecting UV excess met with immediate success, despite some gross initial errors in estimating their numbers on account of confusion with white dwarfs and horizontal-branch stars.1 It quickly became clear that these "radio-quiet" UV-excess objects significantly outnumbered the "radio-loud" quasars by a large factor. Since the word "quasar" derived in part from "radio source," the term "quasi-stellar object (QSO)" was adopted generically for all of these sources. Over time, usage of the term "quasar" has become more generic.

In addition to radio and multicolor optical surveys, low-resolution spectroscopic surveys have been used very effectively to search for AGNs (see Chapter 2 of this volume). Variability, primarily in the optical, has been used to a lesser extent than other methods, but microlensing campaigns, for example, produce large numbers of candidate AGNs as a by-product.

1 The initial overestimate of the space density of quasars was so large that the paper reporting the discovery of quasars by UV excess was entitled "The existence of a new major constituent of the Universe: the quasistellar galaxies" (Sandage 1965).

Rest wavelength (A)

Figure 5.1. Relative energy flux \F\ versus rest-frame wavelength for a composite quasar spectrum. Prominent emission lines are labeled. The lower spectrum, for comparison, is that of an AO v star, in similar arbitrary flux units. The bandpasses for Johnson U and B are also shown. As one goes from longer to shorter wavelengths, note how the A-star spectrum drops dramatically at the Balmer limit, while that of the quasar continues to rise; this accounts for the very blue U — B colors of quasars (see Figure 5.2). The composite spectrum is from the Large Bright Quasar Survey (Francis et al. 1991) and the A-star spectrum is from Pickles (1998). Based on a figure from Peterson (1997).

Rest wavelength (A)

Figure 5.1. Relative energy flux \F\ versus rest-frame wavelength for a composite quasar spectrum. Prominent emission lines are labeled. The lower spectrum, for comparison, is that of an AO v star, in similar arbitrary flux units. The bandpasses for Johnson U and B are also shown. As one goes from longer to shorter wavelengths, note how the A-star spectrum drops dramatically at the Balmer limit, while that of the quasar continues to rise; this accounts for the very blue U — B colors of quasars (see Figure 5.2). The composite spectrum is from the Large Bright Quasar Survey (Francis et al. 1991) and the A-star spectrum is from Pickles (1998). Based on a figure from Peterson (1997).

Was this article helpful?

0 0

Post a comment