The Local Black Hole Mass Function from 2dFGRS Radio Galaxies

Radio emission from early–type (E and S0) galaxies usually signals the presence of a central supermassive black hole. If radio luminosity and black–hole mass are correlated, then we can use the large sample of radio galaxies observed in the 2dF Galaxy Redshift Survey (2dFGRS) to estimate the local (z < 0.1) black–hole mass function for early–type galaxies. Integrating over this mass function yields a local black–hole mass density of ρBH = 1.8+0.4 −0.6 × 10 5 M⊙ Mpc −3 (H0 = 50 km s−1 Mpc,Ω0=1), in good agreement both with the local value derived from galaxy velocity dispersions and with the high–redshift value derived from QSOs. This supports earlier suggestions that local radio galaxies are the direct descendants of most or all of the high–redshift QSOs. 1.1 Does Radio Luminosity Scale with Black Hole Mass? The relationship between black–hole mass and radio luminosity in nearby galaxies is currently a topic of much debate. Auriemma et al. (1977) first showed that (optically) brighter elliptical galaxies are more likely to be radio sources, so the discovery of a correlation between black–hole mass and bulge luminosity (Magorrian et al. 1998) means that some correlation between black–hole mass and radio luminosity must exist for early–type galaxies, at least in a statistical sense (there may still be considerable scatter for individual objects). Franceschini, Vercellone & Fabian (1998) found a remarkably tight relationship between black–hole mass and radio power (Pradio ∝M2.5 BH) in nearby galaxies, implying that the total radio power emitted by a galaxy is both an excellent tracer of the presence of a supermassive black hole and a good estimator of its mass. However, analysis of larger data sets by Laor (2000) and Ho (2002) found a much larger scatter in the MBH–Pradio correlation. All these authors used data sets which contained a mixture of galaxy types (e.g. spirals, ellipticals, Seyfert galaxies, radio galaxies and quasars). More recently, Snellen et al. (2003) have investigated the correlation between black–hole mass and radio power for the large sample of nearby elliptical galaxies with stellar velocity dispersions measured by Faber et al. (1989). They found that that most optically–selected (i.e. ‘relatively passive’) elliptical galaxies obeyed the Franceschini et al. (1998) relation between black–hole mass and radio luminosity. Black hole mass is clearly not the only variable which determines radio luminosity — we