Miguel A. de Avillez

The Generation and Dissipation of Interstellar Turbulence: Results from Large-Scale High-Resolution Simulations

We study, by means of adaptive mesh refinement hydro- and magnetohydrodynamic simulations that cover a wide range of scales (from kiloparsec to subparsec), the dimension of the most dissipative structures and the injection scale of turbulent interstellar gas, which we find to be about 75 pc, in agreement with observations. This is, however, smaller than the average size of superbubbles but consistent with significant density and pressure changes in the ISM, which leads to the breakup of bubbles locally and hence to the injection of turbulence. The scalings of the structure functions are consistent with log-Poisson statistics of supersonic turbulence, where energy is dissipated mainly through shocks. Our simulations are different from previous ones by other authors, since (1) we do not assume an isothermal gas but have temperature variations of several orders of magnitude, and (2) we have no artificial forcing of the fluid with some ad hoc Fourier spectrum but drive turbulence by stellar explosions at the Galactic rate, self-regulated by density and temperature thresholds imposed on the ISM gas.

Testing Global ISM Models: A Detailed Comparison of O VI Column Densities with FUSE and Copernicus Data

We study the O VI distribution in space and time in a representative section of the Galactic disk using three-dimensional adaptive mesh refinement hydrodynamic and magnetohydrodynamic simulations of the interstellar medium (ISM), including the disk-halo-disk circulation. The simulations describe a supernova-driven ISM on large (~10 kpc) and small (~1.25 pc) scales over a sufficiently large timescale (~400 Myr) in order to establish a global dynamical equilibrium. The O VI column density, N(O VI), is monitored through line-of-sight measurements at different locations in the simulated disk. One has been deliberately chosen to be inside of a hot bubble, like our own Local Bubble, while the other locations are random. We obtain a correlation between N(O VI) and distance, which is independent of the observers vantage point in the disk. In particular, the location of the observer inside a hot bubble does not have any influence on the correlation, because the contribution of an individual bubble (with a typical extension of 100 pc) is negligibly small. We find a remarkable agreement between the O VI column densities (as a function of distance) and the averaged O VI density (~1.8 × 10-8 cm-3) in the disk from our simulations and from the values observed with Copernicus and FUSE. Our results strongly support the important role of turbulent mixing in the distribution of O VI clumps in the ISM. Supernova-induced turbulence is quite strong and unavoidable due to shearing motions in the ISM, and it operates on a large range of scales.

MHD Simulations of the ISM: The Importance of the Galactic Magnetic Field on the ISM "Phases"

We have carried out 1.25 pc resolution MHD simulations of the ISM, on a Cartesian grid of 0 ≤ (x, y) ≤ 1 kpc size in the galactic plane and −10 ≤ z ≤ 10 kpc into the halo, thus being able to fully trace the time-dependent evolution of the galactic fountain. The simulations show that large scale gas streams emerge, driven by SN explosions, which are responsible for the formation and destruction of shocked compressed layers. The shocked gas can have densities as high as 800 cm−3 and lifetimes up to 15 Myr. The cold gas is distributed into filaments which tend to show a preferred orientation due to the anisotropy of the flow induced by the galactic magnetic field. Ram pressure dominates the flow in the unstable branch 102 < T ≤ 103.9 K, whereas for T ≤ 100 K (stable branch) magnetic pressure takes over. Near supernovae thermal and ram pressures determine the dynamics of the flow. Up to 80% of the mass in the disk is concentrated in the thermally unstable regime 102 < T ≤ 103.9 K with ∼30% of the disk mass enclosed in the T ≤ 103 K gas. The hot gas in contrast is controlled by the thermal pressure, since magnetic field lines are swept towards the dense compressed walls.

The Distribution of Li-Like Ions in the Local Bubble

We study, by means of three-dimensional adaptive mesh refinement simulations, the spatial distribution of the Li-like ions C IV, N V, and O VI inside the Local Bubble (LB), powered by 19 missing stars of subgroup B1 of the Pleiades suggested as responsible for its origin. The ions C IV, N V, and O VI, and the distribution of their column density ratios (N(C IV)/N(O VI) and N(N V)/N(O VI)) in the simulations are monitored through lines of sight (with lengths as large as 150 pc) emanating from the Solar position. The main results are: (i) there is a weak distribution of C IV and N V in the regions where O VI is strong, (ii) there is a small amount of C IV and N V ions inside the bubble translating into upper limits of –1.0 of the ratios log[N(C IV)/N(O VI)] and log[N(N V)/N(O VI)], consistent with observations, and (iii) with the lack of heat conduction, turbulent mixing, due to the strong shear between the hotter and cooler regions inside the LB, is responsible for the generation and distribution of these ions.

Temperature-averaged and total free-free Gaunt factors for and Maxwellian distributions of electrons ?

Aims. Optically thin plasmas may deviate from thermal equilibrium and thus, electrons (and ions) are no longer described by the Maxwellian distribution. Instead they can be described by

NEI Modelling of the ISM - Turbulent Dissipation and Hausdorff Dimension

\kappa

Dynamical evolution of a supernova driven turbulent interstellar medium

-distributions. The free-free spectrum and radiative losses depend on the temperature-averaged (over the electrons distribution) and total Gaunt factors, respectively. Thus, there is a need to calculate and make available these factors to be used by any software that deals with plasma emission. Methods. We recalculated the free-free Gaunt factor for a wide range of energies and frequencies using hypergeometric functions of complex arguments and the Clenshaw recurrence formula technique combined with approximations whenever the difference between the initial and final electron energies is smaller than

Non-relativistic Free–Free Emission due to the n-distribution of Electrons—Radiative Cooling and Thermally Averaged and Total Gaunt Factors

10^{-10}

Variability of the adiabatic parameter in monoatomic thermal and non-thermal plasmas

in units of

Consequences of Starbursts for the Interstellar and Intergalactic Medium

z^2Ry