Jeffrey S. Oishi

Rapid planetesimal formation in turbulent circumstellar disks

During the initial stages of planet formation in circumstellar gas disks, dust grains collide and build up larger and larger bodies. How this process continues from metre-sized boulders to kilometre-scale planetesimals is a major unsolved problem: boulders are expected to stick together poorly, and to spiral into the protostar in a few hundred orbits owing to a ‘headwind’ from the slower rotating gas. Gravitational collapse of the solid component has been suggested to overcome this barrier. But even low levels of turbulence will inhibit sedimentation of solids to a sufficiently dense midplane layer, and turbulence must be present to explain observed gas accretion in protostellar disks. Here we report that boulders can undergo efficient gravitational collapse in locally overdense regions in the midplane of the disk. The boulders concentrate initially in transient high pressure regions in the turbulent gas, and these concentrations are augmented a further order of magnitude by a streaming instability driven by the relative flow of gas and solids. We find that gravitationally bound clusters form with masses comparable to dwarf planets and containing a distribution of boulder sizes. Gravitational collapse happens much faster than radial drift, offering a possible path to planetesimal formation in accreting circumstellar disks.

YT: A Multi-Code Analysis Toolkit for Astrophysical Simulation Data

The analysis of complex multiphysics astrophysical simulations presents a unique and rapidly growing set of challenges: reproducibility, parallelization, and vast increases in data size and complexity chief among them. In order to meet these challenges, and in order to open up new avenues for collaboration between users of multiple simulation platforms, we present yt (available at http://yt.enzotools.org/) an open source, community-developed astrophysical analysis and visualization toolkit. Analysis and visualization with yt are oriented around physically relevant quantities rather than quantities native to astrophysical simulation codes. While originally designed for handling Enzos structure adaptive mesh refinement data, yt has been extended to work with several different simulation methods and simulation codes including Orion, RAMSES, and FLASH. We report on its methods for reading, handling, and visualizing data, including projections, multivariate volume rendering, multi-dimensional histograms, halo finding, light cone generation, and topologically connected isocontour identification. Furthermore, we discuss the underlying algorithms yt uses for processing and visualizing data, and its mechanisms for parallelization of analysis tasks.

ENZO: AN ADAPTIVE MESH REFINEMENT CODE FOR ASTROPHYSICS

This paper describes the open-source code Enzo, which uses block-structured adaptive mesh refinement to provide high spatial and temporal resolution for modeling astrophysical fluid flows. The code is Cartesian, can be run in one, two, and three dimensions, and supports a wide variety of physics including hydrodynamics, ideal and non-ideal magnetohydrodynamics, N-body dynamics (and, more broadly, self-gravity of fluids and particles), primordial gas chemistry, optically thin radiative cooling of primordial and metal-enriched plasmas (as well as some optically-thick cooling models), radiation transport, cosmological expansion, and models for star formation and feedback in a cosmological context. In addition to explaining the algorithms implemented, we present solutions for a wide range of test problems, demonstrate the codes parallel performance, and discuss the Enzo collaborations code development methodology.

SIMULATING THE COMMON ENVELOPE PHASE OF A RED GIANT USING SMOOTHED-PARTICLE HYDRODYNAMICS AND UNIFORM-GRID CODES

We use three-dimensional hydrodynamical simulations to study the rapid infall phase of the common envelope (CE) interaction of a red giant branch star of mass equal to 0.88 M{sub Sun} and a companion star of mass ranging from 0.9 down to 0.1 M{sub Sun }. We first compare the results obtained using two different numerical techniques with different resolutions, and find very good agreement overall. We then compare the outcomes of those simulations with observed systems thought to have gone through a CE. The simulations fail to reproduce those systems in the sense that most of the envelope of the donor remains bound at the end of the simulations and the final orbital separations between the donors remnant and the companion, ranging from 26.8 down to 5.9 R{sub Sun }, are larger than the ones observed. We suggest that this discrepancy vouches for recombination playing an essential role in the ejection of the envelope and/or significant shrinkage of the orbit happening in the subsequent phase.

Magnetic Fields in Population III Star Formation

We study the buildup of magnetic fields during the formation of Population III star-forming regions by conducting cosmological simulations from realistic initial conditions and varying the Jeans resolution. To investigate this in detail, we start simulations from identical initial conditions, mandating 16, 32, and 64?zones per Jeans length, and study the variation in their magnetic field amplification. We find that, while compression results in some amplification, turbulent velocity fluctuations driven by the collapse can further amplify an initially weak seed field via dynamo action, provided there is sufficient numerical resolution to capture vortical motions (we find this requirement to be 64 zones per Jeans length, slightly larger than but consistent with previous work run with more idealized collapse scenarios). We explore saturation of amplification of the magnetic field, which could potentially become dynamically important in subsequent, fully resolved calculations. We have also identified a relatively surprising phenomenon that is purely hydrodynamic: the higher-resolved simulations possess substantially different characteristics, including higher infall velocity, increased temperatures inside 1000?AU, and decreased molecular hydrogen content in the innermost region. Furthermore, we find that disk formation is suppressed in higher-resolution calculations, at least at the times that we can follow the calculation. We discuss the effect this may have on the buildup of disks over the accretion history of the first clump to form as well as the potential for gravitational instabilities to develop and induce fragmentation.

Turbulent Torques on Protoplanets in a Dead Zone

Migration of protoplanets in their gaseous host disks may be largely responsible for the observed orbital distribution of extrasolar planets. Recent simulations have shown that the magnetorotational turbulence thought to drive accretion in protoplanetary disks can affect migration by turning it into an orbital random walk. However, these simulations neglected the disks ionization structure. Low ionization fraction near the midplane of the disk can decouple the magnetic field from the gas, forming a dead zone with reduced or no turbulence. Here, to understand the effect of dead zones on protoplanetary migration, we perform numerical simulations of a small region of a stratified disk with magnetorotational turbulence confined to thin active layers above and below the midplane. Turbulence in the active layers exerts decreased, but still measurable, gravitational torques on a protoplanet located at the disk midplane. We find a decrease of 2 orders of magnitude in the diffusion coefficient for dead zones with dead-to-active surface density ratios approaching realistic values in protoplanetary disks. This torque arises primarily from density fluctuations within a distance of one scale height of the protoplanet. Turbulent torques have correlation times of only ~0.3 orbital periods and apparently time-stationary distributions. These properties are encouraging signs that stochastic methods can be used to determine the orbital evolution of populations of protoplanets under turbulent migration. Our results indicate that dead zones may be dynamically distinct regions for protoplanetary migration.

Cassiopeia A and Its Clumpy Presupernova Wind

The observed shock wave positions and expansion in Cassiopeia A can be interpreted in a model of supernova interaction with a freely expanding stellar wind with a mass loss rate of ~2 × 10-5 M☉ yr-1 for a wind velocity of 10 km s-1. The wind was probably still being lost at the time of the supernova, which may have been of Type IIn or Type IIb. The wind may play a role in the formation of very fast knots observed in Cas A. In this model, the quasi-stationary flocculi (QSFs) represent clumps in the wind, with a density contrast of several 103 compared with the smooth wind. The outer, unshocked clumpy wind is photoionized by radiation from the supernova and is observed as a patchy H II region around Cas A. This gas has a lower density than the QSFs and is heated by nonradiative shocks driven by the blast wave. Denser clumps have recombined and are observed as H I compact absorption features toward Cas A.

ON HYDRODYNAMIC MOTIONS IN DEAD ZONES

We investigate fluid motions near the midplane of vertically stratified accretion disks with highly resistive midplanes. In such disks, the magnetorotational instability drives turbulence in thin layers surrounding a resistive, stable dead zone. The turbulent layers in turn drive motions in the dead zone. We examine the properties of these motions using three-dimensional, stratified, local, shearing-box, non-ideal, magnetohydrodynamical simulations. Although the turbulence in the active zones provides a source of vorticity to the midplane, no evidence for coherent vortices is found in our simulations. It appears that this is because of strong vertical oscillations in the dead zone. By analyzing time series of azimuthally averaged flow quantities, we identify an axisymmetric wave mode particular to models with dead zones. This mode is reduced in amplitude, but not suppressed entirely, by changing the equation of state from isothermal to ideal. These waves are too low frequency to affect sedimentation of dust to the midplane, but may have significance for the gravitational stability of the resulting midplane dust layers.

The Inability of Ambipolar Diffusion to Set a Characteristic Mass Scale in Molecular Clouds

We investigate the question of whether ambipolar diffusion (ion-neutral drift) determines the smallest length and mass scales on which structure forms in a turbulent molecular cloud. We simulate magnetized turbulence in a mostly neutral, uniformly driven, turbulent medium, using a three-dimensional, two-fluid, magnetohydrodynamics (MHD) code modified from Zeus-MP. We find that substantial structure persists below the ambipolar diffusion scale because of the propagation of compressive slow MHD waves at smaller scales. Contrary to simple scaling arguments, ambipolar diffusion thus does not suppress structure below its characteristic dissipation scale, as would be expected for a classical diffusive process. We have found this to be true for the magnetic energy, velocity, and density. Correspondingly, ambipolar diffusion leaves the clump mass spectrum unchanged. Ambipolar diffusion appears unable to set a characteristic scale for gravitational collapse and star formation in turbulent molecular clouds.

MAGNETOROTATIONAL TURBULENCE TRANSPORTS ANGULAR MOMENTUM IN STRATIFIED DISKS WITH LOW MAGNETIC PRANDTL NUMBER BUT MAGNETIC REYNOLDS NUMBER ABOVE A CRITICAL VALUE

The magnetorotational instability (MRI) may dominate outward transport of angular momentum in accretion disks, allowing material to fall onto the central object. Previous work has established that the MRI can drive a mean-field dynamo, possibly leading to a self-sustaining accretion system. Recently, however, simulations of the scaling of the angular momentum transport parameter αSS with the magnetic Prandtl number Pm have cast doubt on the ability of the MRI to transport astrophysically relevant amounts of angular momentum in real disk systems. Here, we use simulations including explicit physical viscosity and resistivity to show that when vertical stratification is included, mean-field dynamo action operates, driving the system to a configuration in which the magnetic field is not fully helical. This relaxes the constraints on the generated field provided by magnetic helicity conservation, allowing the generation of a mean field on timescales independent of the resistivity. Our models demonstrate the existence of a critical magnetic Reynolds number Rmcrit, below which transport becomes strongly Pm-dependent and chaotic, but above which the transport is steady and Pm-independent. Prior simulations showing Pm dependence had Rm < Rmcrit. We conjecture that this steady regime is possible because the mean-field dynamo is not helicity-limited and thus does not depend on the details of the helicity ejection process. Scaling to realistic astrophysical parameters suggests that disks around both protostars and stellar mass black holes have Rm Rmcrit. Thus, we suggest that the strong Pm dependence seen in recent simulations does not occur in real systems.