Simulations from Waters & Proga (2016)

On the efficient acceleration of clouds in active galactic nuclei
(Submitted to MNRAS Letters)
Astro-ph submission

1-D Simulations

Part 1: cloud formation

We first evolved the solutions from Proga & Waters (2015) until the cloud was fully formed. Averages (over hot and cold gas) of the radiation force and primitive variables (density, velocity, and pressure) are plotted as a function of time on the left panels. The grey vertical line marks the time corresponding to the spatial profiles (right side panels).

Part 2: cloud acceleration

The above configuration serves as our initial conditions. We now introduce a time-varying ionizing flux and show that this leads to more efficient cloud acceleration. The constant flux solution, CF1D, is shown using dotted lines. Notice that the solution is perfectly cyclic after about 40 thermal times.

2-D Simulations

Part 1: cloud formation

In 2D, we again evolve the cloud for 25 thermal times keeping the flux constant, at which point the cloud is already disrupted by the Rayleigh-Taylor instability. The grey vertical line now marks the time corresponding to the density maps (bottom images).

Part 2: cloud acceleration

Upon introducing the time-varying flux, we again find that this leads to more efficient cloud acceleration. The constant flux solution is denoted CF2D and its temporal properties are shown by the dotted lines. The net acceleration of run CF2D is just 2-3 times that of gravity, while for the variable flux solution (run VF2D), the factor is 6.5.



These movies are available for download here:
1Dpart1, 1Dpart2, 2Dpart1, 2Dpart2.
(You probably need to right-click and select "Save Link As...")

Acknowledgements

This work was supported by NASA under ATP grant NNX14AK44G. Our simulations were performed on the Eureka cluster at UNLV's National Supercomputing Center for Energy and the Environment (NSCEE). Software used for this work: simulations: Athena v4.2; visualization and analysis: SciPy and yt.