Novak – “Hydrodynamic simulations of black hole fueling and feedback”

• “Nuclear physics”: not analytic, non-local, inhomogeneous, coupled at vastly different length scales, energy generation affects fueling
• Axi-symmetric simulations: 2.5pc to 250kpc scales, run for Gyrs.  See his papers in the last few years. Physically rich model (RT, feedback, dust models & opacity, etc.)
• Increasing the dimensionality of the simulations smoothes out the accretion rate. Why? Turbulent cascade behaviors in 2D (amplified into blobs -> big accretion events) vs. 3D (cascades into smaller blobs -> smooth accretion).
• Working on a subgrid model for AGN: time-averaged behavior as a function of smoothing radius (so it can be applied to various sim. setups). Inside ~200 pc, accretion rates don’t depend on density, sound speed, or BH mass! Outside, the sound speed dependence goes negative (c_s^β, β < 0). Outside 1 kpc, increasing dependence on density (ρ^α, α > 0).

Nagamine – “Gas accretion onto a supermassive black hole: a step to model AGN feedback”

• Bridging the gap between <pc and ~kpc simulations.
• Assumptions in cosmo simulations: Bondi accretion, radiative output from accretion (L = ε * Mdot * c²), radiative efficiency, feedback efficiency.
• Non-spherical in/outflows develops from radiative feedback through thermal instabilities.
• AGN feedback efficiencies in their sims are low (<1e-4)…
• Q&A session: Steve Allen found that the Bondi rate (from temperature and density from X-ray observations) matched well with the observed jet power.

Park – “Accretion onto black holes from large scales regulated by radiative feedback”

• Explored parameter space (average Mdot, bursty accretion, duty cycle) in RT 2D simulations.
• The mean period between bursts is proportional to the sound crossing time across the Strömgren sphere.
• Currently, exploring RT instabilities in I-fronts. If R_s ~ R_B, then the HII region collapses.

Parkin – “Magnetized accretion disk turbulence”

• Convergence in magnetic energy input / injection on large scale.  Requires increased resolution?
• Do BCs matter? Shearing box vs. global disk (but not 2π).
• They find convergence in global disk sims (alpha ~ 0.04) and lower resolution sims (~30 cells/H) in their global sims vs. shearing boxes.