Accretion onto the supermassive black hole in M87

Chandra X-ray observations of the giant elliptical galaxy M87 resolve the thermal state of the hot interstellar medium into the accretion (Bondi) radius of its central 3 x 10(9) M-circle dot black hole. We measure the X-ray gas temperature and density profiles and calculate the Bondi accretion rate, (M)over dot(Bondi) similar to 0.1 M-circle dot yr(-1). The X-ray luminosity of the active nucleus of M87 observed with Chandra is L-X,L-0.5-7 keV similar to 7 x 10(40) ergs s(-1). This value is much less than the predicted nuclear luminosity, L-Bondi similar to5 x 10(44) ergs s(-1), for accretion at the Bondi rate with a canonical accretion radiative efficiency of 10%. If the black hole in M87 accretes at this rate it must do so at a much lower radiative efficiency than the canonical value. The multiwavelength spectrum of the nucleus is consistent with that predicted by an advection-dominated flow. However, as is likely, the X-ray nucleus is dominated by jet emission then the properties of flow must be modified, possibly by out flows. We show that the overall energetics of the system are just consistent with the predicted Bondi nuclear power. This suggests that either most of the accretion energy is released in the relativistic jet or the central engine of M87 undergoes on-off activity cycles. We show that, at present, the energy dumped into the ISM by the jet may reduce the accretion rate onto the black hole by a factor proportional to (v(j)/c(s))(-2), where v(j) is the jet velocity and c(s) the ISM sound speed, and that this is sufficient to account for the low nuclear luminosity.