11.4. AGN Classification: QI, QII and QIII

Despite the large diversity, all AGNs seem to share some common properties. It is therefore useful to attempt to construct a detailed classification scheme, related to as many as possible observed characteristics. The main division I propose is to three luminosity classes: QI, QII and QIII. They overlap, in luminosity, with what is sometimes called Quasars, Miniquasars (Seyfert 1 and bright Seyfert 2 galaxies) and Microquasars (faint Seyferts and LINERs). The subdivision is into spectral types, using the intensity ratio of the broad and narrow component of the permitted lines. This follows a suggestion by Osterbrock and collaborators to assign a number between 1.0 and 2.0 to each Seyfert galaxy, according to its narrow/broad line ratio. Objects of type 1.0 do not show any narrow permitted lines and objects of type 2.0 have only narrow lines. The narrow hydrogen lines of many Seyfert 1 galaxies are 10-20% as intense as the broad lines. These have been classified as Seyfert 1.5 galaxies. As for NLXGs and LINERs that show faint broad H wings, they were classified as Seyfert type 1.8 or 1.9. Thus a simple way to classify Seyfert galaxies, using the broad and narrow Balmer lines, is

(104)

Other properties are self-explanatory and are summarized in Table 3.

Table 3. AGN classification

	QI	QII	QIII
Luminosity (X+ optical)	1045-47erg s-1	1043-45erg s-1	1041-43erg s-1
Spectral type	1.0 - 1.5	1.0 - 2.0	1.8 - 2.0
Line width(km s-1)	400 - 10,000	300 - 10,000	200 - 400
continuum variability	yes	yes	not observed
line variability	in some	in most	not observed
line or continuum reddening	?	in some	in most
[OIII]5007 / H(narrow)	6-12	6-12	1-2
[OII]3727 / [OIII]5007	0.1-0.3	0.1-0.6	1-5
[NII]6583 / H(narrow)	0.5-1	0.3-0.8	0.8-2
[OI]6300 / H(narrow)	0.05-0.3	0.05-0.3	0.1-0.6
[SII]6716 + 6731 / H(narrow)	0.3-0.7	0.3-0.7	0.5-1.5
EW(H)	~ 100Å	10 - 100Å	1 -10Å
Ionization parameter (BLR)	0.03 - 1	0.03 - 1	?
Ionization parameter (NLR)	~ 10-2	10-2-10-3	10-3-10-4

The table refers to observed properties, and some of the entries are still questionable. For example, the fraction of very bright objects (QI) showing any indication of reddening is not known, for reasons explained in chapter 7 (the entry in the table refers to the observed property of extinction and not to dust emission). Also, the intrinsic continuum luminosity of some QII objects (the Seyfert 2s and the NLXGs ) may be much larger than indicated by their classification.

Some of the AGN properties seem to change, continuously, along the QI, QII, QIII sequence. The most obvious ones are:

The spectral type: The relative strength of the narrow emission lines increases along the sequence.

Reddening: Line and continuum reddening is more noticeable in low luminosity AGNs.

The NLR ionization parameter: Accepting the idea of photoionization as the main excitation mechanism for all classes of AGNs, the narrow emission lines seem to form a single parameter sequence, with the ionization parameter decreasing from QI to QIII.

Variability: Much more common in QI, although the observations of the non-stellar continuum, and the broad lines in the QII and QIII classes leave much to be desired.

There are glaring omissions too, most notably the central mass and accretion rate. Adding these is a real challenge for future AGN studies.

Obviously, some of the above properties, and trends, are subject to ambiguity and selection effects. Nevertheless, the connection between the sub-classes starts to reveal itself, and we are perhaps justified in looking for a common physical mechanism in all AGNs.