Methods based on gas kinematics

Unlike the situation for stars where anisotropy clouds the interpretation, gas kinematics is much easier to interpret if the gas is in Keplerian motion^11.13. There are two caveats, however. There may be non-gravitational forces, magnetic fields for example. Second, there is no a priori reason for the gas to be in Keplerian motion. So we need to lucky then.

Optical emission lines Several examples of nuclear disks around centres of galaxies were found by HST. For example for M87 (the big elliptical in the Virgo cluster), HST measurements found a Keplerian rotation curve in the disk with a velocity of 1000km s$^{-1}$ at a distance of 19 from the centre^11.14. This implies a MDO with mass $\approx 2.4\times 10^9\hbox{$M_\odot$}$. Similar evidence for a MDO now exists for other galaxies as well.

Maser emission Some AGN have luminous maser sources close to the centre. Radio observations of these maser lines show that the masers are in nearly Keplerian rotation, and from the speed one can place tight constraints on the mass density of the MDO, $> 5\times 10^{12}\hbox{$M_\odot$}$.

Reverberation mapping Another elegant way to determine the extent of an AGN is `reverberation mapping'. Many of the emission lines seen from an AGN are due to the light from the central source being reprocessed by the surrounding material. As the central source varies in luminosity, the emission lines follow, but with a time-lag that corresponds to the light-travel time between the central source and the line-emitting gas. So by measuring the time-lag, we can estimate the size of the light-emitting region.