Radiative Models of Sagittarius A* and M87 from Relativistic MHD Simulations
Abstract
The discovery that the magnetorotational instability (MRI) is likely the
mechanism of angular momentum transport and accretion has led to rapid
progress in the theory of black hole accretion. General relativistic
MHD (GRMHD) simulations currently provide the most
realistic physical description of black hole accretion flows, but
neglect radiation and are currently only applicable to low luminosity
systems, where the radiation can be added in post-processing. To
connect simulations with observations, we have developed codes
for the semi-analytic computation of Kerr photon orbits
(\texttt{geokerr}) and for the calculation of time-dependent, general
relativistic radiative transfer via ray tracing (\texttt{grtrans}).
The Galactic center black hole candidate, Sgr A*, provides an
ideal first problem for study using GRMHD and these radiative transfer
methods. Recent very long baseline
interferometry (VLBI) observations detected the structure of its
accretion flow on event horizon scales, allowing a direct comparison
between accretion disk theory and observation. We compare millimeter
wavelength images and light curves from a set of three-dimensional
GRMHD simulations to VLBI and spectral observations of Sgr A* using
general relativistic radiative transfer. The GRMHD models provide an
excellent fit to current observations. Magnetic turbulence driven by
the MRI naturally explains the observed
millimeter variability. Fitting models to observations allows
estimates of the inclination and position angles of the black hole, as
well as of the median electron temperature in the millimeter emission
region and the accretion rate onto the black hole. The black hole
shadow, a signature of the event horizon, is unobscured in all models
and may be detectable with VLBI experiments in the next 5-10 years. We
also consider ``tilted" accretion disk models, where the angular
momentum axis of the accretion flow is misaligned from the black hole
spin axis. The additional degree of freedom in these models allows a
wide range of viable possibilities for Sgr A*. Additional observations
both of visibility amplitudes and closure phases will constrain the
models further. We also study generic observational consequences of
tilt in black hole accretion disks, and find that the radiation edge
of these systems is independent of black hole spin; in stark contrast
to the untilted disks where it decreases with increasing spin, roughly tracking the marginally stable
orbit. Line profiles based on gravitational redshifts and Doppler
shifts of these simulations vary strongly with observer
azimuth. Coupled with precession, this may lead to strongly
time-varying emission lines from tilted, geometrically thick disks.
Finally, radiative models of M87 are
constructed in a similar fashion to Sgr A*. The M87 spectrum can be
fit by either a jet or a jet/disk model. The geometry in M87 can be
reasonably constrained by observations of the extended jet emission,
and the resulting millimeter images are fairly robust despite
considerable model uncertainties. The Gaussian FWHM inferred from mm-VLBI
on current telescopes should be 36-41 $\mu$as. Jet images are more
compact than disk images, and in either case the shadow in M87
should be accessible to future VLBI observations.
Collections
- Physics [291]